Nvidia is set to launch an AI hub in Stockholm, Sweden, aimed at assisting startups in addressing health and environmental issues. 

This initiative sees Nvidia teaming up with Norrsken, a non-profit organization founded by Niklas Adalberth, who is also a co-founder of the Swedish payment platform Klarna. Norrsken actively supports startups focused on tackling social and ecological challenges.

Named the Stockholm AI Lab, this hub is expected to open its doors in the heart of Stockholm by 2027, as shared in Norrsken’s recent announcement. The facility will house selected startups alongside Nvidia employees and feature a biotechnology demonstration lab.

Nvidia’s involvement will include providing essential computing infrastructure, software tools, and access to its global network of startups, although specific investment details have yet to be revealed.

What to Expect Inside the Hub

The Stockholm location will offer private office spaces and serve as a venue for events designed to foster connections among founders, investors, researchers, industry leaders, and policymakers. Niklas Adalberth emphasized the necessity of a collaborative space, stating that founders need a dedicated environment to connect with potential investors and clientele and to foster a sense of community in their endeavors.

The demonstration lab will be managed by Nvidia in collaboration with Root Bio, a Stockholm-based startup that utilizes AI for molecular design relevant to biological applications, including efforts to lower greenhouse gas emissions. The initiative aims to showcase how artificial intelligence can facilitate advancements in biological discovery, with plans to include additional partners focused on robotics, climate initiatives, health, and manufacturing in the future.

Nvidia’s software will provide foundational support for projects across robotics, simulation, and life sciences, although the announcement did not clarify the types of equipment to be installed, the computing capacity available to members, or any associated costs for these resources.

A Broader European Initiative

This new hub in Stockholm is part of Nvidia’s larger strategy within Europe to empower companies and researchers in utilizing AI technology, which manifests in various forms across the continent.

For instance, in September 2025, Nvidia unveiled a robotics research and development facility in the UK, collaborating with techUK, a technology trade organization, along with QA to cultivate AI and robotics skills. Additionally, in June 2025, Nvidia announced the establishment or expansion of several AI technology centers in countries such as Germany, Italy, Spain, Finland, Sweden, and the UK, furthering research, training, and access to AI computing resources.

Nvidia is launching an AI hub in Stockholm for start-ups addressing health and climate issues.

A bright, sunlit co-working space inside a sleek Stockholm glass hub, natural morning light flooding through floor-to-ceiling windows overlooking the city skyline. Entrepreneurs at worktables study large touchscreens: one shows a DNA helix seamlessly transforming into green leaves beside climate data charts. A living vertical garden climbs one wall. A single Swedish flag stands by the entrance, and a bold Nvidia logo is mounted prominently on an interior wall.

Nvidia joins Klarna co-founder’s foundation to launch Stockholm AI hub - Euronews.com

- Nvidia is opening an AI hub in Stockholm to address health and climate challenges.  
- The hub will partner with Norrsken and host start-ups alongside a demonstration lab.  
- Nvidia aims to facilitate collaboration between founders, investors, and industry stakeholders.  

Nvidia is opening an AI hub in Stockholm, Sweden, to help start-ups tackle health and climate challenges.

ADVERTISEMENT ADVERTISEMENT

The US chipmaker is partnering with Norrsken, a Swedish non-profit foundation established by Niklas Adalberth, co-founder of Swedish payments company Klarna. The foundation supports start-ups working on social and environmental problems.

Called Stockholm AI Lab, the hub is due to open in central Stockholm in 2027, according to Norrsken’s announcement.

It will bring selected start-ups and Nvidia staff under the same roof, alongside a biotechnology demonstration lab.

Nvidia will contribute computing infrastructure, software tools and connections to its international start-up network. The partners have not disclosed the size of the investment in the project.

What will happen inside?

The hub in central Stockholm will provide private offices and host events connecting founders with investors, researchers, industry representatives and policymakers.

“What these founders need now is what every great movement needs: a place. Somewhere to find each other, their investors, their first customers, and the conviction that they are not building alone,” Adalberth said in the announcement.

The demonstration lab will be run by Nvidia and Stockholm-based Root Bio, a start-up using AI to design molecules for biological applications, including reducing greenhouse gas emissions.

The partners say it will show how AI can help automate biological discovery. They also plan to select further demonstration partners working in robotics, climate, health and manufacturing.

Nvidia’s software tools will support work in robotics, simulation and life sciences

However, the announcement does not specify which equipment will be installed, how much computing capacity members will receive or what they will pay to use it.

Part of Nvidia’s wider European push

Nvidia has announced similar efforts elsewhere in Europe to help businesses and researchers use AI, although the projects take different forms.

In the United Kingdom (UK), the company announced a robotics research and development hub with technology trade association techUK in September 2025, working with training company QA to develop AI and robotics skills.

In June 2025, Nvidia also announced new or expanded AI technology centres in Germany, Italy, Spain, Finland, Sweden and the UK, supporting research, training and access to AI computing infrastructure.

Nvidia to open Stockholm AI hub for start-ups tackling health and climate challenges

Hex-Rays, a Belgian company renowned for its reverse-engineering software IDA Pro, is set to acquire the Israeli startup HexAI, formerly known as Ventris, in an estimated deal between $30 million and $35 million. Founded in 2025 by the 25-year-old Elon Gliksberg, HexAI has worked quietly on a pioneering operating system aimed at equipping AI agents for security research tasks. The startup, which consists of a small team of six, has not sought external funding, and Hex-Rays intends to inject resources into enhancing its presence in Israel.

Elon Gliksberg's journey into programming and cybersecurity began in middle school, where he initially created hacks for video games. During high school, he developed multiple applications for his peers, including a ride-sharing app reminiscent of Uber, which helped students organize transportation after getting their licenses. His technical journey continued in Israel's Unit 8200, where he took on leadership roles for various projects.

Post-military service, Gliksberg joined a startup as its first employee, aiming to gain insights into the industry before venturing into his own enterprise. He has authored several technical papers and conducted vulnerability research that uncovered significant security weaknesses in services run by major corporations, earning acknowledgments and rewards for his disclosures.

With the growing use of AI agents, Gliksberg recognized a significant opportunity to leverage these technologies for security research, leading to the establishment of HexAI. The goal was to enhance the measurability of the research process and ensure that the results could be independently verified. Amit Schendel, who had previously worked alongside Gliksberg in Unit 8200, later took on the role of Chief Technology Officer at HexAI.

HexAI caters to both governmental and commercial clients, both in Israel and abroad. The company has reported that clients benefit from not only an increase in the overall quality and volume of security findings but also significant reductions in the costs associated with AI models. Notably, these improvements can be achieved by utilizing openly available models rather than depending on premium systems from companies such as OpenAI and Anthropic.

The platform empowers security researchers to design and run their research workflows, evaluate their efficacy, and authenticate their results. It specializes in analyzing both source code and binary files to uncover known vulnerabilities, including those associated with Common Vulnerabilities and Exposures (CVE) as well as zero-day vulnerabilities and malware.

Researchers can customize workflows to meet their specific needs, evaluate their accuracy, and continually refine their approaches. In addition to user-generated workflows, HexAI provides pre-built options that undergo ongoing development and testing against industry standards and common goals in security research. Some workflows are even capable of autonomously enhancing other research processes.

Gliksberg emphasized, “Rather than depending on generic agents that function independently, which can be pricey and inconsistent, our users can concentrate on their specific research objectives, assess performance, and validate outcomes. They have the flexibility to create workflows tailored to their requirements, dictate agent functionality, and iterate on results. Furthermore, they can utilize our pre-existing workflows that our research teams continually refine without needing to construct everything from scratch.”

The platform is optimized for extensive research initiatives and efficient use of AI tokens. It possesses the versatility to function in air-gapped environments that keep systems isolated from outside networks—an invaluable asset for organizations engaged in sensitive security research that cannot be shared with external services.

Through acquiring HexAI, Hex-Rays aims to broaden its capabilities in AI-driven security research and tackle a wider array of cybersecurity challenges. With 35 years of expertise in reverse engineering, Hex-Rays is well-positioned to utilize its flagship product, IDA Pro, which allows researchers to analyze software even when the source code is not accessible.

25-year-old Israeli entrepreneur divests stealth AI company for potentially $35 million without revealing details.

A sunlit Tel Aviv rooftop office with golden afternoon light filtering through glass walls. A confident young adult in smart casual attire shakes hands with a suited investor across a sleek table where a single AI chip prototype rests beside a discreet stack of cash. The Israeli flag hangs softly in the background, warm tones highlighting their faces and the sense of a stealth startup’s quiet triumph.

25-year-old Israeli founder sells stealth AI startup for up to $35 million without raising capital - www.calcalistech.com

- Hex-Rays is acquiring Israeli startup HexAI for $30 million to $35 million.  
- HexAI develops an AI operating system for security research, serving global customers.  
- The acquisition aims to enhance Hex-Rays' cybersecurity offerings and AI capabilities.  

Hex-Rays, the Belgian developer of the reverse-engineering tool IDA Pro, is acquiring Israeli startup HexAI (previously Ventris) in a deal valued at $30 million to $35 million. Founded in 2025 by 25-year-old Elon Gliksberg, HexAI has remained in stealth while developing an operating system for AI agents designed to conduct security research. The six-person company did not raise outside capital, and Hex-Rays plans to invest in expanding its operations in Israel.

Gliksberg began programming and exploring cybersecurity in middle school, when he wrote cheats for video games. In high school, he developed several applications used by classmates, including an Uber-style app that allowed students to coordinate rides after obtaining their driver's licenses. He later served in several roles in Israel's Unit 8200, where he led projects.

1 View gallery Elon Gliksberg. ( Photo: HexAI )

After completing his military service, Gliksberg joined a startup as its first employee to gain experience before launching his own company. He published technical articles and vulnerability research, identified critical security flaws in services operated by multibillion-dollar companies, and received compensation for reporting them.

As the use of AI agents expanded, Gliksberg identified an opportunity to apply them to security research. He founded HexAI to make the process measurable and its findings verifiable. Amit Schendel, who served alongside Gliksberg in Unit 8200, later joined the company as CTO.

HexAI serves government and commercial customers in Israel and internationally. The company says its customers have reported improvements in both the quality and quantity of security findings, alongside substantial reductions in AI model costs. Those gains can be achieved using open models, without relying on the most advanced systems from providers such as OpenAI and Anthropic.

The platform allows security researchers to build and run research workflows using AI agents, measure their performance and validate their findings. It analyzes source code and binary files to identify known vulnerabilities, including those associated with Common Vulnerabilities and Exposures (CVE) identifiers, as well as zero-day vulnerabilities, malware and backdoors.

Researchers can create customized workflows, assess their accuracy and refine them. HexAI also offers prebuilt workflows that its research teams continuously develop and test against benchmarks and common security research targets. The platform includes workflows designed to improve other research workflows autonomously.

“Instead of relying on generic agents operating on their own, which are expensive and unreliable, our users can focus on what they want to research, measure performance and validate findings,” Gliksberg said. “They can build research workflows around their needs, define how the agents work, and test and improve the results. They can also run built-in workflows that our research teams continuously test and improve, without having to build everything themselves.”

The system is designed for large-scale research and efficient AI token usage. It is not tied to a specific AI model and can operate in air-gapped environments, where systems are isolated from external networks. Such capabilities can be particularly relevant to organizations conducting sensitive security research that cannot be transferred to external services.

Hex-Rays is acquiring HexAI to expand into AI-powered security research and address a broader range of cybersecurity challenges. The Belgian company has 35 years of experience in reverse engineering, and its flagship product, IDA Pro, enables researchers to analyze software even when its source code is unavailable.

25-year-old Israeli founder sells stealth AI startup for up to $35 million without ra

Disney is making a significant investment by onboard 66 employees from Character.AI, an AI chatbot company that has recently declared bankruptcy, shortly after reducing its own technology workforce. According to a Chapter 11 filing from Character.AI's parent company, Disney is set to allocate $60 million annually for these new hires. This move follows Disney's appointment of Character.AI's CEO, Karandeep Anand, as its first tech chief in mid-September, although specific details regarding Anand's compensation remain undisclosed.

This strategic hiring initiative aligns with Disney’s broader AI strategy under the leadership of CEO Josh D'Amaro, who has been channeling resources into enhancing the company's technological platforms. Disney's stock has experienced a modest increase of less than 2% over the past year and a significant decline of over 36% in the last five years, prompting these tech investments.

The decision to bring in Character.AI personnel comes just weeks after Disney let go of several hundred employees, including some from its tech divisions. This downsizing marked the third significant round of layoffs enacted by D'Amaro since he took charge in mid-March.

In conjunction with the $60 million earmarked for employee salaries, Disney is also establishing a $30 million "aggregate retention pool" to encourage these new hires to remain with the company. The total expenditure for compensation and retention over the upcoming three years is projected to be around $180 million, as indicated in the filing.

Among the 66 employees offered positions at Disney, 56 have already exited Character.AI, while the remaining 10 are awaiting updates on their work visas from the U.S. government. Entertainment lawyer Corey Martin commented on the high compensation figures, noting that Character.AI's workforce represents its most valuable asset. "While the numbers may seem startling, they reflect the standard in this industry," Martin said, clarifying that he has no involvement in the deal or with Disney.

Additionally, Disney agreed to pay $20 million for key assets belonging to Character.AI, which includes intellectual property and technology. Competing bids for these assets will be accepted until November 20, with oversight provided by a bankruptcy court judge. Disney has not responded to inquiries regarding this matter.

Character.AI entered bankruptcy amid lawsuits from families alleging that their teenagers experienced severe distress or even suicide after interacting with the company’s AI chatbots. Earlier this year, the startup and Google, one of its investors, reached settlements regarding several child safety lawsuits.

The company had received a cease-and-desist notice from Disney last fall, prohibiting the creation of AI chatbots that feature Disney's copyrighted characters without explicit permission. Character.AI complied with this request by removing chatbot functionalities associated with popular franchises like Star Wars and Marvel.

In its bankruptcy statement, Character.AI cited "substantial legal, regulatory, and operational challenges." In a memo to Disney employees, Anand expressed enthusiasm for the new role, stating that the company is "uniquely positioned to thrive" in a competitive landscape that is currently experiencing pivotal changes. He lauded the talent and dedication of the Disney team, indicating excitement about fostering a collaborative environment moving forward.

Disney Compensates 66 Character.AI Employees with $60 Million Following Tech Job Cuts

An upscale corporate boardroom bathed in warm morning sunlight, a diverse group of adult professionals gathered around a long polished wooden table. At center lies an oversized ceremonial check made out for “$60,000,000 to 66 Character.AI Staffers.” Behind them, a sleek glass wall features a large Disney logo. Scattered coffee cups and paperwork hint at recent layoff turmoil, while relieved smiles and supportive pats on shoulders capture the moment of overdue compensation.

Disney is paying $60 million to hire 66 Character.AI staffers - Business Insider

- Disney plans to pay $60 million yearly for 66 employees from Character.AI.  
- The company is setting aside $30 million for a retention pool to keep staff.  
- Character.AI filed for bankruptcy amid lawsuits related to its custom AI chatbots.  

Disney is paying up for employees from an AI chatbot startup that filed for bankruptcy, shortly after trimming its own tech workforce.

The Mouse House plans to pay $60 million per year to 66 workers from Character.AI, according to a Chapter 11 filing on Tuesday from its parent company. In mid-September, Disney hired Character.AI CEO Karandeep Anand to be its first-ever company-wide tech chief. (The company hasn't yet disclosed how much it's paying Anand.)

These additions are part of Disney's AI strategy under CEO Josh D'Amaro, who's been investing in the media giant's tech platforms to help jumpstart a stagnant stock that's up less than 2% in the last 12 months and is down over 36% in the last five years.

Disney's decision to bring in these Character.AI staffers came two weeks after it laid off several hundred workers — including some tech employees. The late-September cuts were Disney's third major round of layoffs since D'Amaro took over in mid-March.

In addition to the $60 million in compensation, Disney is setting aside an additional $30 million for an "aggregate retention pool" to incentivize Character.AI employees to stay. The company expects to spend $180 million on "total compensation and retention" for these staffers in the next three years, the filing said.

Related video Amazon cuts 14,000 corporate jobs amid AI restructuring

Of the 66 Character.AI staffers who were offered jobs by Disney, 56 have already resigned from the AI startup, while the other 10 are waiting for the US government to update their work visas.

Entertainment lawyer Corey Martin of Granderson Des Rochers, who advises companies on media transactions, said he's not surprised by those pay figures since Character.AI's employees are its biggest asset.

"They're highly compensated employees. The numbers can be eye-popping, but it's not that unusual in this space," said Martin, who didn't work on the deal or with Disney.

Disney also agreed to pay $20 million for Character.AI's key assets, including some of its IP and tech. Other companies can submit competing bids for the assets by November 20. The entire process will be overseen by a bankruptcy court judge. Disney didn't respond to requests for comment for this story.

Character.AI filed for bankruptcy after facing lawsuits from families who alleged that their teenagers harmed themselves or committed suicide after using the company's custom AI chatbots. The AI startup and Google, one of its backers, settled several child safety lawsuits earlier this year.

Disney sent a cease-and-desist letter to Character.AI last fall, telling it not to create AI chatbots featuring its copyrighted characters without permission. Character.AI complied, removing chatbots with characters from Disney franchises like Star Wars and Marvel.

Explore BI Games Take a smarter break in your day - and see how far you get. Play now Want more Business Insider in your news feed? Add BI in Google so our reporting is easier to find when you’re searching for what matters. Add to Preferred Sources

In its bankruptcy announcement, Character.AI said that it had been facing "significant legal, regulatory, and operational headwinds."

Anand told Disney employees in a memo last Friday that the Mouse House was "uniquely positioned to win" in a competitive media business that he said was "at an inflection point." The tech chief also said he enjoyed meeting his new colleagues and looked forward to working with them.

"What stands out most is the talent and care this team puts into its work, the craft Disney has perfected, and the strong foundation you've built," Anand said.

Disney Pays $60 Million to 66 Character.AI Staffers After Tech Layoffs

In a recent addition to The Indicator from Planet Money, the focus shifts to Character.ai, an AI startup that finds itself in a precarious position as it files for Chapter 11 bankruptcy. This development invites a closer examination of the implications of 'reverse acquihires' in the tech landscape.

The company, known for its ability to engage users with AI-powered companions—ranging from fictional characters to historical figures—gained significant traction after launching shortly before ChatGPT, amassing 28 million monthly active users at its peak, partially due to its popularity on TikTok. In a bold move, Google acquired the founders and the nonexclusive rights to their technology for $2.7 billion in 2024, opting not to integrate Character.ai into its operations, but rather to use the founders for other projects. This strategy is termed a 'reverse acquihire' and has emerged as a trend among AI startups, with at least six similar incidents occurring over the last two years.

However, the aftermath of Google's involvement was not as rosy for Character.ai as one might expect. Following the acquisition, the company had to pivot from its expensive large language model (LLM) training to focus on developing AI characters, all while leadership faced challenges in monetizing their existing user base. Though it appeared financially sound with $2.7 billion in the bank, the departure of its founders left the remaining team struggling to regain footing.

The circumstances surrounding Character.ai's bankruptcy raise questions about whether the reverse acquihire model was directly responsible for its struggles or merely a reflection of more significant issues. The startup grappled with legal challenges concerning alleged harm to minors and ongoing copyright disputes with Disney. The outcome? A once-promising venture now finding itself subject to Disney's scrutiny for parts of its technology, valued at just $20 million.

In other news, President Trump has entered into an agreement to import diesel fuel from Russia, aimed at reducing fuel prices ahead of the midterm elections. This unexpected move is drawing bipartisan criticism, especially in light of recent sanctions aimed at curtailing Russia's energy revenues. Senators have urged Trump to reconsider, citing concerns that this arrangement could inadvertently support Russia's military actions in Ukraine.

Although Goldman Sachs suggests this deal may only minimally affect wholesale prices—projected to drop between $1 and $2 per barrel—analysts believe it will not substantially shift the current dynamics of the diesel market. Adding to the mystery, Russia has faced its own challenges, including damaged refineries and the need to import diesel to meet domestic shortages, raising questions about the viability of this deal.

On the academic front, if you haven't heard about Michael Woodford and Robert Hall, who recently won the Nobel Prize in Economics for shaping New Keynesian economics, it’s worth a listen. Their work intertwined microeconomic concepts with traditional Keynesian frameworks. Notably, the Nobel Prize originally did not include economics—an omission rectified by the Swedish Central Bank for its 300th anniversary by establishing an award just for that purpose.

As the week unfolds, these narratives intertwine the worlds of economic theory, AI innovation challenges, and geopolitical considerations, shaping the larger discourse on technology and its impact in today's world.

Implications of this bankruptcy for AI startups

A dark, empty AI startup office at night: a broken “AI Startup” neon sign flickers above glass doors plastered with a yellow “BANKRUPTCY NOTICE,” inside desks are strewn with abandoned laptops showing paused code and half-drunk coffee cups, dim server racks hum in the gloom, shadows stretching across empty chairs—evoking the collapse of an AI venture.

What this bankruptcy says about AI startups - NPR

- Character.ai, an AI startup, has filed for Chapter 11 bankruptcy this week.  
- The company struggled after a "reverse acquihire" and changes in its business model.  
- Litigation over copyright and harm to minors contributed to Character.ai's financial troubles.

What this bankruptcy says about AI startups

You're reading a newsletter from The Indicator from Planet Money. Once a week, we curate our favorite stories and insights on business, finance, economics and why they matter. Subscribe here to get it in your inbox every Friday.

After you win the economics Nobel, what would your first act be? Buy a Corvette? Gloat to your childhood bully? For co-winner Michael Woodford, his first act was … proctor an exam. The DAY that Columbia professor Michael Woodford found out that he and Robert Hall won the economics Nobel, he returned to his class with the pithy title "Cognitive Mechanisms and Economic Behavior" to administer an exam. He was welcomed to the classroom with cheers and applause. He held up his arms in triumph.

toggle caption i Andrew Harrer/Bloomberg via Getty Images

Coming up:



Character.ai is going through bankruptcy. What can we learn from the AI startup's decline?

Why President Trump wants to import diesel fuel from Russia

Why stagflation was John Maynard Keynes' Achilles' heel

News We're Watching

When AI startups get vacuumed up, what's left behind?

What to know: Character.ai is a company that allows you to speak with a "companion" — an AI-powered fictional character, historical figure, or whatever you decide to build. It came out just before ChatGPT and quickly went viral in part thanks to its adoption on TikTok. It hit 28 million monthly active users at its peak.

Sponsor Message

Character.ai had a lot of engagement, a big customer base, lots of talent … and Google promptly picked 'em up for $2.7 billion back in 2024. But Google didn't buy the company and absorb it! No, that traditional "acquihire" model would require all sorts of legal and regulatory approvals. Instead, they hired away the founders (and other big hitters on the team) and bought the nonexclusive license to the technology, leaving the company itself behind. That's called a "reverse acquihire," and it's happened at least six times in the past two years with AI startups. Now this week, Character.ai is filing for Chapter 11 bankruptcy.

toggle caption i Gabby Jones/Bloomberg via Getty Images

Why it matters: A "reverse acquihire" doesn't leave a company in the greatest position. After the Google deal, Character.ai had to change its business model away from the quite costly LLM training and toward developing AI characters. Its founders were gone. Sure, the shrunken company had $2.7 billion in the bank, but leadership was still looking to raise money as they struggled to monetize their user base.

So is the "reverse acquihire" what caused the pain? Or was it a symptom of a larger issue? Character.ai faced a mountain of litigation over alleged harm to minors plus copyright concerns from Disney. So what's left behind? A once very promising company, which Disney is very excited to pick over for parts of it for $20 million.

Sponsor Message

Will this Russian diesel deal affect price?

What's going on: President Trump signed a deal last week to get diesel from Russia in an effort to lower fuel prices ahead of the midterms. Two weeks ago, we talked in the newsletter about the potential of a diesel export ban to lower prices. It didn't happen. One week ago, we talked again about Trump issuing an executive order to defer a tax from diesel. Now, this!

It's safe to say people were surprised by the move, given Trump signed a Russian sanctions law just last month that looked to tamp down on their energy revenues. Beyond surprise, a bipartisan group of senators is openly calling for the president to rescind the deal. They said it gives Putin the funds to power his war machine and "enables Russia to continue its brutal attacks on Ukrainian civilians."

Will this do anything? The same bipartisan group of senators said this deal will mean "limited to no relief to Americans." Goldman Sachs estimated the release of Russian oil would affect wholesale prices by $1 to $2 per barrel. The price of a barrel of diesel is around $200. So a $1 to $2 impact? Neil Atkinson, an independent energy analyst, says, "Frankly, [it's] a rounding error. It makes no fundamental difference to the global diesel market."

What's weirder to Atkinson, though … Russia hasn't really been exporting diesel. Its refineries have been damaged, and it's literally been importing diesel to make up for shortfalls. " So where these barrels of Russian diesel are going to come from to go either to the United States or elsewhere in the global market is a bit of a mystery," said Atkinson.

Today's Indicator

The Nobel Prize that wasn't

If you HAVEN'T listened: Reporter Greg Rosalsky opens the piece describing WHAT exactly Michael Woodford and Robert Hall won the economics Nobel for. They helped shape what's known as NEW Keynesian economics by bringing microeconomic principles to the traditional models of John Maynard Keynes. BUT THEN, Greg goes on to explain that the Nobel Prize did not initially include an economics award. No, that changed when the Swedish Central Bank, the oldest central bank in the world, wanted to celebrate its 300th anniversary with a bang. They would fund a new award for economics. It's a fascinating story which you can listen to below.

Sponsor Message

If you HAVE listened: A leading philosophy of John Maynard Keynes is that governments could fight recessions by boosting spending, and deal with inflation by holding spending back. In the initial draft that didn't make it into the final episode, Greg touched on how STAGFLATION was a central thorn in the side of this concept. Specifically, back in the 1970s, during the Arab oil embargo, the U.S. dealt with the one-two punch of high unemployment and high inflation, bringing about stagflation. It challenged the Keynesian idea that when lots of people are out of work and spending less, inflation comes down. Because at that point, unemployment and inflation rose together. Isn't that nice? Greg Rosalsky writes Planet Money's newsletter … you can catch that here.

Rest of the week



In Our Spare Time

Producer Cooper Katz McKim is reading The AGI Chronicles

Man oh man, this is the page-turniest nonfiction book I've read in a while. Kevin Roose, co-host of the upcoming NPR podcast Machine Gods, formerly The New York Times' Hard Fork, tracks the handful of characters racing to make AI a commercial reality. It is far from a clear-cut story. There's no one character who came up with the idea and executed it … the existence of AI today comes down to many papers, many individual discoveries, and a whole lot of obsession. I didn't realize how much progress had been made in AI prior to 2020 and how concerned many industry leaders were about even creating AI (looking at you, Amodei) for fear of creating a super-intelligence that moves outside of human control. It's gossip-y, it's matter of fact … I would recommend.

What this bankruptcy says about AI startups

Ramp, a company specializing in expense management solutions, announced on Thursday that it successfully secured $1.85 billion in its latest funding round, elevating its valuation to $60 billion. This comes just four months after it raised $750 million, which valued the company at $44 billion.

The funding round was co-led by prominent investment firms including Dragoneer Investment Group, Thrive Capital, Founders Fund, Sands Capital, and Avenir Growth.

The company reported a significant increase in its annualized revenue, which has surpassed $2 billion, a rise from $1.5 billion noted in June. This growth trajectory is 25% faster than the growth rate recorded during the same period last year. Ramp now facilitates an impressive annualized purchase volume exceeding $275 billion, reflecting a 175% increase compared to the previous year. The startup has grown its customer base to 85,000, up from 70,000 in June, including high-profile clients such as Visa, Uber, Shopify, Robinhood, Anduril, Notion, Cursor, and Figma.

This new funding reflects Ramp's efforts to broaden its services beyond traditional expense management. The company is now venturing into payments, fraud detection, procurement, vendor management, and recently, accounting. 

As part of its forward-thinking approach, Ramp is integrating artificial intelligence into its offerings by embedding AI agents in its various services, including procurement and budget management. It has introduced a corporate credit card tailored for AI agent utilization and launched a model routing service designed to engage with the growing need for AI inference management, which complements its existing product suite that features AI token usage monitoring and token spend management.

Furthermore, Ramp recently unveiled Glass, a workspace dedicated to collecting and retaining AI automations, prompts, and skills developed by its employees, enabling a repository of organizational knowledge for collective use.

Christian Jensen, a partner at Dragoneer, emphasized Ramp's evolution into an extensive financial operations platform, stating, “In just a few years, Ramp has become a much broader financial operations platform, with AI deeply embedded in how the company builds and operates. We believe Ramp is well on its way to becoming one of the defining companies of this era.”

The investment round also welcomed participation from other notable investors, including Capital International, Fidelity Management & Research Company, T. Rowe Price, D1 Capital Partners, Greenoaks, and many more. In total, Ramp has garnered approximately $5 billion in funding.

As Ramp expands its range of products, it faces increasing competition from Rippling, which also offers HR and payroll software along with corporate cards and expense management solutions, leading to a direct rivalry over similar finance team budgets.

Keith Rabois, an early investor in Ramp, expressed confidence in the company’s direction during a recent interview, stating that while competition from Rippling exists, the primary goal is to rival industry giants like Anthropic. “We’re going to compete with Anthropic... I want to be larger and more successful and more important than Anthropic,” Rabois noted, alluding to Ramp’s ambition to establish itself as the leading platform for managing token expenditures and optimizing return on investment.

Ramp achieves $60 billion valuation as it expands further into AI.

Sunlit modern office lobby with a gigantic Ramp logo in brushed metal on a marble wall; in front, a warm‐gold digital display graph soars past “$60B” alongside stylized neural‐network patterns; soft natural light floods the scene, and a lone executive silhouette studies the AI-enhanced data feed.

Ramp reaches $60B valuation as it pushes deeper into AI - TechCrunch

- Ramp raised $1.85 billion, increasing its valuation to $60 billion in four months.  
- Annual revenue surpassed $2 billion, with a 25% growth rate compared to last year.  
- The company is expanding into various services, including AI-enhanced expense management and procurement.  

Expense management company Ramp said on Thursday that it had raised $1.85 billion in a new funding round that values it at $60 billion, just four months after raising $750 million at a $44 billion valuation.

The financing was co-led by Dragoneer Investment Group, Thrive Capital, Founders Fund, Sands Capital, and Avenir Growth.

Ramp said its annualized revenue has crossed $2 billion, up from a reported $1.5 billion in June, and that its revenue growth is 25% faster than it was around this time last year. The company claims it now facilitates annualized purchase volume of over $275 billion, up 175% from a year earlier. It also says it has 85,000 customers, up from 70,000 in June, including Visa, Uber, Shopify, Robinhood, Anduril, Notion, Cursor, and Figma.

The funding comes as Ramp has been expanding beyond expense management into payments, fraud detection, procurement, vendor management, and, of late, even accounting.

The startup has also built an AI story around itself, embedding AI agents into its procurement, expense management, accounting, budgeting, and other products. It also launched a corporate credit card specifically designed for AI agents to use, and in August entered the model routing business to tap the booming AI inference market and give existing clients a model routing service that fits neatly with its current products, including AI token usage monitoring and token spend management.

Earlier this month, Ramp launched Glass, a workspace that collects and remembers the AI automations, prompts, and skills that an organization’s employees build so others can take advantage of the company’s collective knowledge and skills.

“In just a few years, Ramp has become a much broader financial operations platform, with AI deeply embedded in how the company builds and operates. We believe Ramp is well on its way to becoming one of the defining companies of this era,” Christian Jensen, partner at Dragoneer, said in a statement.

Other investors in the round include Capital International, Fidelity Management & Research Company, T. Rowe Price, D1 Capital Partners, Greenoaks, ICONIQ, GIC, 8VC, 1789 Capital, Alpha Wave Global, Ontario Teachers’ Pension Plan, Coatue, D.E. Shaw, Definition Capital, FirstMark, General Catalyst, Generation Investment Management, Glade Brook Capital Partners, Khosla Ventures, Lightspeed Venture Partners, Pinegrove Opportunity Partners, Soma Capital, and Stripes.

The company has now raised about $5 billion in funding.

As Ramp has widened its product lineup, it has increasingly run into Rippling, the HR and payroll software company that has pushed into corporate cards, expense management, and bill payments. The overlap puts the two companies in direct competition for the same finance-team budgets.

Early investor Keith Rabois, who has backed Ramp through two firms, Founders Fund and Khosla Ventures, both of which participated in this round, spoke to TechCrunch last month about Ramp’s strategy and his own big ambitions for the company.

“Ramp is not worried about Rippling,” Rabois said. “We’re going to compete with Anthropic,” he continued. “That’s the goal. I want to be larger and more successful and more important than Anthropic. Ramp is going to be the primary way people manage their token spend, how they optimize it, how they measure ROI, and that will lead us to collide probably with Anthropic in many ways.”

— With additional reporting from Connie Loizos

Ramp reaches $60B valuation as it pushes deeper into AI

In a growing number of high-tech workplaces, artificial intelligence agents are beginning to outnumber human workers, creating new challenges for oversight. Opsin, a startup focused on AI security and governance, announced a successful $21 million funding round on Thursday, driven by the increasing presence of AI tools from firms like OpenAI and Microsoft within corporate environments. This surge in AI utilization raises concerns about the potential risks posed by these advanced technologies.

With the deployment of autonomous software that performs various tasks becoming accessible even to those without technical backgrounds, there's a risk that employees may lose track of what these AI agents are actually doing in their organizations. James Pham, CEO and cofounder of Opsin, noted that security apprehensions were hindering AI adoption, prompting the need to expand his company's capabilities.

"Agents are more akin to humans than you might assume," Pham commented. "While we may not fully comprehend what goes on in a human's mind, their actions provide insight into their behavior."

Opsin's mission is to deliver a detailed understanding of AI agents' activities within a business and assess any associated risks. The funding round, led by SYN Ventures and Race Capital, aims to bolster the startup's sales and engineering departments.

The rise of AI agents has left many security professionals grappling with the complexities of this emerging landscape. Pham, an MIT graduate, teamed up with cofounder Oz Wasserman after both left Abnormal Security in 2022. They consulted extensively with numerous stakeholders to grasp the hesitations surrounding AI integration into workplaces, with a notable concern being the potential misuse of sensitive data. 

Opsin now provides tools to track AI agents' operations, detailing which data they access and their intended functions while identifying those that may pose risks. The company's services include ongoing oversight of these agents to highlight any behaviors that could lead to security issues. "Ultimately, it's all about the users' behaviors and how the agents behave," Pham emphasized.

The sheer number of AI agents in even less tech-oriented firms has astonished some clients, as demonstrated by a case in which a CFO's AI agent unintentionally granted other employees elevated access to confidential financial records.

As AI technology continues to proliferate, the cybersecurity sector is experiencing a significant upsurge in activity. Glenn Chisholm, a cyber investor supporting Opsin, explained that the increasing use of AI products opens up more avenues for cyberattacks. "The more individuals who engage with these systems, the easier it becomes for attackers to infiltrate and exploit them," he noted.

Interest in cybersecurity stocks has surged this year, and venture capital investments are on the rise, with AI firms like Anthropic and OpenAI actively expanding their hiring in cybersecurity roles. Opsin aims to enhance the safety of organizations utilizing tools like Microsoft Copilot and ChatGPT Enterprise. Chisholm remains optimistic about the industry's future, asserting that the next few years will involve capitalizing on AI's capabilities rather than fearing its implications.

Opsin Raises $21 Million to Address AI Agents in the Corporate Sector

A sunlit modern boardroom with a large Opsin logo etched on a glass wall; a confident executive in a sharp suit gestures at a digital display where stylized AI agent icons—small robot silhouettes—are being guided into a secure vault graphic. Warm natural light floods the polished conference table, hinting at growth and a $21 M expansion.

This startup raised $21 million to clean up the AI agent mess hitting companies - Business Insider

- Opsin raised $21 million to help monitor and manage AI agents in workplaces.  
- AI agents can potentially misuse sensitive data, raising significant security concerns for companies.  
- The cybersecurity industry is adapting rapidly as AI usage in businesses increases.  

AI agents are starting to outnumber humans in many high-tech workplaces, and they can be far trickier to monitor.

Opsin, an AI security and governance startup, announced Thursday it raised $21 million to meet the moment. As employees across the corporate world plug AI agents from the likes of OpenAI and Microsoft into more of their work, the company is helping to wrangle the new crop of potentially risky interlopers.

It's never been easier for non-technical workers to deploy agents — autonomous software built on AI that can complete tasks. As employees pump the software into their companies' work, code, and databases, they can lose sight of what the AI is actually doing.

When Opsin's CEO and cofounder, James Pham, realized security fears were slowing down AI use, he decided it was time to scale up his company, he told Business Insider.

"Agents are actually more similar to humans than you think," Pham said. "And the reason is because you don't understand what is going on in a human head, but you can judge them by their actions."

Related video Sam Altman explains why AI will replace 40% of your work

Opsin aims to provide a comprehensive view of what AI agents are actually doing within an organization and the risks that behavior might pose. SYN Ventures and Race Capital led the roughly 20-person company's new funding round, which will go toward expanding its sales and engineering teams.

Security pros need to clean up a mess of AI agents

Pham, an MIT alum, and one of his cofounders, Oz Wasserman, met at the firm Abnormal Security and each left in 2022. When they decided to launch a company, Pham said they spoke with hundreds of people to understand the concerns about bringing AI into businesses.

The issue of sensitive data leaped out. Pham said that person after person raised concerns that AI agents might access and misuse personal or confidential information.

Now, Opsin helps clients show where agents are active, which tools and data they access, what they were originally created to do, and which ones might need cleaning up. Then, the firm offers to continuously monitor companies' work, watch over agents, and flag risky behavior.

"At the end of the day, it's all about the behaviors of the users, the behavior of the agent," Pham said.

He said the scale of agents, even in relatively less tech-savvy companies, has stunned some customers. At one firm Opsin worked with, an agent made by a chief financial officer inadvertently gave other employees top-level access to financial information, Pham said.

Explore BI Games Take a smarter break in your day - and see how far you get. Play now Want more Business Insider in your news feed? Add BI in Google so our reporting is easier to find when you’re searching for what matters. Add to Preferred Sources

AI's cyber threats could mean a boom

The cybersecurity industry has been in a frenzy, and now the growth of agents is adding to that storm. Glenn Chisholm, a cyber investor backing Opsin with SYN Ventures, told Business Insider that as companies use more AI products, hackers will find more targets.

"The more people that are doing it, the easier it is for an attacker to gain access and purchase," Chisholm said.

Cyberdefense stocks have soared this year, and Chisholm said he's seen more venture capital investment. Anthropic and OpenAI are each hiring aggressively in cybersecurity.

Opsin hopes it can make companies feel safer using Microsoft Copilot, ChatGPT Enterprise, and other products. Chisholm is optimistic that the industry will adapt to AI.

"The next couple of years are us working out how to get the most out of it, rather than sort of fearing the concept of it," he said.

Opsin Expands With $21M to Tackle AI Agents in Businesses

Booking Holdings is initiating a new AI travel startup named Vayo in Seattle, with seasoned entrepreneur Mike Fridgen at the helm. The company plans to invest up to $100 million based on Vayo's development progress. Designed to elevate the travel experience, Vayo's AI technology scans users' emails to extract confirmations for flights, hotels, and restaurants while monitoring for price reductions and availability at fully booked accommodations. The startup views general-purpose AI platforms, like Meta’s Muse, as its primary competition, focusing on a niche solution tailored exclusively for travel.

This venture signifies a comeback for Booking Holdings in Seattle, the city that houses Expedia Group, its primary competitor. The company previously established a presence in Seattle through Booking.com in 2012, relocated to Bellevue the following year, and ultimately closed the office in 2020 during the pandemic.

In other corporate news, Microsoft shares reached an all-time high, reflecting a remarkable surge that has added approximately $1.4 trillion to its market value since late June—equivalent to about half of Amazon’s worth in less than four months. This rally follows a challenging first half of the year, which saw Microsoft suffer a $357 billion loss in a single day. A significant contributor to the turnaround was Azure achieving over $100 billion in annual revenue as of July.

Meanwhile, Bellevue-based Humanly (humanly.io) has acquired CareerArc, merging its automated candidate screening tools with CareerArc's resources to streamline the hiring process for employers, enhancing the journey from initial application to final interview.

In economic news, a recent survey conducted by the Seattle Metropolitan Chamber of Commerce indicates a decline in economic optimism among voters, with 68% reporting a belief that the local economy is deteriorating. Despite a temporary boost from this summer’s FIFA World Cup, concerns regarding business retention, rising local costs, and public safety are mounting, particularly among major tech companies.

Getty Images has announced it will maintain its Seattle headquarters and avoid layoffs while navigating bankruptcy to restructure debt following a failed merger with Shutterstock. The Getty family and Koch Equity Development will inject as much as $533 million into the company as it faces challenges from the increasing prevalence of AI-generated imagery.

In the gaming sector, Microsoft Xbox CEO Asha Sharma is launching Xbox Labs to explore innovative ideas, including game remixes and adaptations to new hardware, even if these concepts challenge existing business models. Scott Van Vliet, the Xbox CTO, will oversee the new initiative.

In civic engagement, Steve Ballmer’s USAFacts hosted a YouTube debate featuring conservative commentator Ben Shapiro and progressive author Scott Galloway. Under Ballmer's “Fact Desk,” the debate centered on whether free markets or government policies are more effective at addressing the cost-of-living crisis and maintaining the American Dream.

In tech developments, Amazon's latest Alexa Tablet 12 Pro has been reviewed, highlighting its full Android capability and an impressive anti-glare display, though it has faced criticism for having excessive built-in bloatware. 

Microsoft has also addressed over 1,400 security vulnerabilities, surpassing its prior record of nearly 1,000 in September, with AI playing a crucial role in identifying many of these issues. Security experts warn that this high volume is putting pressure on corporate security teams testing updates.

In partnership news, Expedia has entered a multiyear agreement to become the National Hockey League’s official travel partner in North America, promoting its brand during both regular-season and playoff games.

Additionally, Workday is acquiring C1, previously known as ConductorOne, a Portland-based identity security startup that manages employee and AI agent access, underlining the growing need for control over AI functionalities in business.

The University of Washington is leading an initiative that has received $20 million from the National Science Foundation to enhance an AI Institute focused on creating open-source software for modeling complex systems.

Lastly, Seattle’s Porch Group is moving forward with plans to purchase Allied Trust Insurance Company, further extending its offerings in home service software and insurance.

Booking Holdings is launching a new AI travel company in Seattle.

A bright, sunlit innovation hub in downtown Seattle at sunrise: floor-to-ceiling windows frame the Space Needle and city skyline. Inside, a sleek glass wall bears a large Booking Holdings logo. On a wooden conference table, an open laptop displays a dynamic world map with glowing flight routes. Natural light pours in over green plants and modern furnishings, capturing the excitement of a new AI‐driven travel startup.

Booking Holdings is building a new AI travel startup in Seattle - LinkedIn

- Booking Holdings is launching an AI travel startup called Vayo with a $100 million investment plan.  
- Microsoft’s stock reached a record high, adding $1.4 trillion in market value since June.  
- Getty Images will retain its Seattle HQ during bankruptcy, with investments from the Getty family and Koch.

Booking's new Seattle startup: Booking Holdings (NASDAQ: BKNG) is launching an AI travel venture called Vayo , led by veteran Seattle entrepreneur Mike Fridgen , with plans to invest $100 million depending on its progress. Vayo's AI agents pull flight, hotel and restaurant confirmations from a user's email, then watch for price drops, openings at sold-out hotels and other changes. The startup sees general-purpose AI agents such as Meta's Muse as its main competition, betting that agents built only for travel will do better.

Vayo marks a return to Seattle for Booking Holdings in the home city of its biggest rival, Expedia Group . Booking.com opened an outpost in the city in 2012, moved it to Bellevue the following year, and closed it in 2020, at the height of the pandemic.

Microsoft bounces back: Microsoft ’s stock closed at a record high Tuesday, capping a run that has added about $1.4 trillion in market value since late June — the equivalent of roughly half an Amazon in less than four months. The rally follows a rough first half, when Microsoft lost $357 billion in a single day in January. Azure topping $100 billion in annual revenue in July helped spark the turnaround.

Humanly (humanly.io) makes another acquisition: The Bellevue-based startup is adding CareerArc to its growing roster, merging candidate-attraction tools with its automated screening AI to give employers a more unified pipeline from first click to final interview.

Post-World Cup reality check: A new Seattle Metropolitan Chamber of Commerce survey reveals that voters' economic optimism has plummeted, with 68% saying the city’s economy is getting weaker. While this summer's FIFA World Cup provided a brief economic high, voters are increasingly anxious about business retention, local costs, and public safety as major tech employers continue to balance their regional footprints.

Getty Images’ headquarters at 605 Union Station in Seattle. (GeekWire Photo / Kurt Schlosser)

Getty survives in Seattle: Getty Images plans to keep its Seattle HQ and avoid layoffs as it goes through bankruptcy to cut its debt, after a failed merger with Shutterstock deepened its financial troubles. Members of the Getty family and Koch Equity Development will invest up to $533 million, taking joint ownership of the stock photo giant as it grapples with the rise of AI imagery and other long-term headwinds.

Xbox embraces the remix: Microsoft Xbox CEO Asha Sharma is creating a team called Xbox Labs to experiment with game remixes, ports to new hardware and AI agents, even when those ideas challenge the company’s own business. Xbox CTO Scott Van Vliet will lead the group.

Data meets debate: Steve Ballmer ’s civic data initiative USAFacts hosted a YouTube debate pitting conservative commentator Ben Shapiro against progressive author and professor Scott Galloway. With the former Microsoft CEO running the "Fact Desk," the showdown addressed whether free markets or government policies are better equipped to tackle the cost-of-living crisis and preserve the American Dream.

Hot Links

Amazon's Alexa Tablet 12 Pro finally runs full Android and features a "gorgeous" anti-glare display, according to The Verge's review. The biggest drawback: built-in bloatware, including two fixed homescreen widgets that largely serve as ads for Amazon products. (The Verge)

Microsoft fixed more than 1,400 security flaws on Tuesday, topping September's record of nearly 1,000, crediting AI for finding many of the bugs. Security experts warn the volume is straining corporate security teams that test updates before rolling them out. (Krebs on Security)

Expedia is hitting the ice as the National Hockey League (NHL) ’s official travel partner in North America, under a multiyear deal that puts its brand on the boards during regular-season and playoff games. (PSBJ)

Workday is buying C1 , formerly ConductorOne, an identity security startup with a significant Portland presence that manages access for employees and AI agents. The deal underscores the need to control what AI agents can do as they become more common in business. (Workday)

The AI Institute in Dynamic Systems, a University of Washington -led program including multiple institutions, received $20 million from the National Science Foundation to continue building open-source software to model complex systems. (UW)

Seattle-based Porch Group has announced plans to acquire Allied Trust Insurance Company as it continues expanding its home services software and insurance offerings. (Business Wire)

Sign up here to receive GeekWire's daily newsletter in your inbox each day at 11 a.m. PT.

Booking Holdings is building a new AI travel startup in Seattle

The United Kingdom is rapidly harnessing the full potential of technology across various sectors, driving significant advancements from established enterprises to innovative startups tackling some of the world's toughest challenges. 

A key player in this evolution is NVIDIA, which has committed $250 million over five years to enhance research in safety and security within the U.K. This move reinforces the company's dedication to supporting the local startup ecosystem, contributing to businesses such as Basecamp Research, Ineffable Intelligence, Synthesia, Wayve, among others.

NVIDIA recently unveiled plans to provide additional funding to companies like Boltz, Foresight, Hiverge, Prima Mente, and Riverlane. Boltz, established in London by MIT researchers, focuses on developing biomolecular models and AI agents that assist scientists in predicting molecular structures and interactions, thereby aiding in the design of new medications. 

Foresight, another London startup, specializes in helping large infrastructure project owners anticipate completion dates, mitigate risks, and expedite revenue generation. Meanwhile, Hiverge, founded by researchers from Google DeepMind and the University of Cambridge, is innovating an AI discovery engine that automatically generates and refines algorithms for tasks like AI training acceleration and complex logistics. 

Prima Mente is dedicated to creating biological foundation models aimed at unraveling intricate diseases and pinpointing new therapeutic targets, initially concentrating on conditions such as Alzheimer’s, Parkinson’s, and ALS. Riverlane, based in Cambridge, is advancing quantum computing through real-time error correction hardware and software that accommodates all major qubit types, making quantum processors more reliable for practical applications.

In addition to these investments, NVIDIA is joining forces with the Earthshot Prize as a Global Alliance Founding Partner, applying its computing and technical resources to assist innovators in addressing global climate and sustainability issues.

In the energy sector, Fuse Energy is deploying its customer AI on an in-house infrastructure powered by NVIDIA's B300 Blackwell Ultra, resulting in improved support for over 450,000 households across the U.K. Simultaneously, Locai Labs, a sovereign AI enterprise, has launched Nemotron-Personas-UK, an open-source dataset featuring one million synthetic personas that highlight the U.K.'s demographic, regional, and cultural diversity, aiding developers in crafting more equitable AI systems.

In healthcare, OneAdvanced is rolling out what it claims is the country’s premier fine-tuned large language model (LLM) for AI triage in primary care—built using NVIDIA’s Agent Toolkit software and open models—benefitting millions of NHS patients. 

Heidi has integrated NVIDIA's Nemotron open models to expedite access to trusted medical information for clinicians. In another significant advancement, Speechmatics is employing the multilingual medical speech recognition model Oak with NVIDIA’s Triton Inference, allowing healthcare providers to control processing of sensitive patient audio. 

Additionally, Apian, Synnovis, and NVIDIA are collaborating to utilize advanced computing and digital simulations via NVIDIA Cosmos, enabling hospital robots to manage the transportation of lab samples, allowing medical personnel to devote more attention to patient care. Danaher is also working with NVIDIA to enhance autonomous scientific methods, integrating accelerated computing and AI to transform laboratory workflows. Their newly announced AI-powered lab aims to quicken the discovery of affinity reagents using NVIDIA's BioNeMo models.

On the civic front, the Liverpool City Region is developing an NVIDIA-supported "City Brain," a sovereign AI initiative serving as a template for various Unitary Authorities throughout the U.K. Additionally, Digital Jersey, Innovation City Belfast, and Tees Valley Combined Authority are investigating how trustworthy AI computing can benefit their communities and public services.

In the financial sector, LSEG and NVIDIA are collaborating to implement GPU-accelerated market analytics for global organizations. Earlier this year, Revolut launched PRAGMA, a foundational model for banking event sequences trained on 26 million anonymized user histories, transitioning from research to real-time application.

Lloyds Banking Group developed ATUM, a foundation model designed to translate billions of transactions into reusable AI, while Starling is crafting a banking foundation model hosted on Civo's U.K. sovereign cloud infrastructure. NatWest is partnering with NVIDIA on Nemotron Voice to improve understanding of how U.K. banking customers communicate, and Modulr is deploying a network of agents powered by Nemotron for various payment operations.

As the quantum computing landscape evolves, U.K. enterprises like Riverlane are proving the potential to integrate classical AI with quantum systems. The company has demonstrated that GPUs can expedite decoder recalibration processes essential for quantum error correction, aligning decoder performance with real-time changes in quantum hardware capabilities. This work is part of a larger initiative to explore synergies among AI, accelerated computing, and quantum technologies to support future advancements.

These developments illustrate the thriving innovation ecosystem in the U.K., with over 3,100 companies participating in NVIDIA's Inception program aimed at promoting cutting-edge startups. For further insights, tune into Jensen Huang's keynote at the GTC Berlin event on October 21 at 11 a.m. CEST.

NVIDIA Unveils New Investments in the UK Startup Sector

A sunlit London co-working loft with floor-to-ceiling windows overlooking the Shard and Tower Bridge. A diverse group of adult entrepreneurs stands around a high table strewn with laptops and tiny hardware boards, marveling at a small glowing green “circuit-board sapling.” On the exposed brick wall behind them, a large NVIDIA logo glows in its signature green, and a single UK flag hangs nearby.

NVIDIA Announces New Investments in the UK Startup Ecosystem - NVIDIA Blog

- NVIDIA is investing $250 million in the U.K. for safety and security research initiatives.  
- New partnerships support AI startups focusing on healthcare, quantum computing, and infrastructure projects.  
- Civic and financial sectors are leveraging NVIDIA's technology to enhance public services and analytics.  

Across industries, the U.K. is putting the full stack to work.

The scale of deployment is growing fast, from enterprises transforming their businesses to startups solving some of the planet’s hardest problems.

NVIDIA is putting that principle into practice, including through a five-year, $250 million compute commitment to strengthen the U.K.’s agentic safety and security research.

NVIDIA’s Investments in the UK

NVIDIA has been investing in the U.K. startup ecosystem, supporting companies including Basecamp Research, Ineffable Intelligence, Synthesia, Wayve and many more.

Today, NVIDIA announced new planned investments in Boltz, Foresight, Hiverge, Prima Mente and Riverlane.

Boltz, a London-based AI startup founded by MIT researchers, builds biomolecular models and AI agents that help scientists predict molecular structures and interactions, and design potential new medicines.

Foresight, a London-based startup, helps owners of major infrastructure projects, including hyperscale data centers, predict completion dates, mitigate risks and accelerate time to revenue.

Hiverge, a Cambridge-based AI startup founded by researchers from Google DeepMind and the University of Cambridge, builds an AI discovery engine that automatically writes and optimizes algorithms for tasks such as faster AI training and complex logistics planning.

Prima Mente, a London-based AI biology startup, focuses on building biological foundation models to decode complex diseases and identify novel therapeutic targets, starting with Alzheimer’s, Parkinson’s and ALS.

Riverlane, a Cambridge-based quantum computing startup, builds real-time error correction hardware and software that work across all major qubit types, helping quantum computers become reliable enough for practical use.

In addition, NVIDIA is partnering with the Earthshot Prize, joining as a Global Alliance Founding Partner to bring compute and technical expertise to innovators working on solutions for the world’s greatest climate and sustainability challenges.

Technology Leaders Pioneer Intelligent Solutions Across the UK

In energy, Fuse Energy is moving its customer AI onto its own in-house infrastructure, built on NVIDIA B300 Blackwell Ultra, delivering faster, more responsive support to over 450,000 U.K. households.

Meanwhile, sovereign AI company Locai Labs released Nemotron-Personas-UK — an open source dataset of one million synthetic personas representing the UK’s demographic, regional and cultural diversity — helping developers build fairer, more representative AI systems.

Healthcare

In healthcare, OneAdvanced is deploying what it calls the U.K.’s first and most advanced fine-tuned LLM for Primary Care AI-triage — built with NVIDIA Agent Toolkit software and open models — supporting millions of National Health Service patients.



Heidi has now integrated NVIDIA Nemotron open models to give clinicians faster access to trusted medical evidence.

Speechmatics is deploying Oak, a multilingual medical speech recognition model, using NVIDIA Triton Inference — giving healthcare organizations control over where sensitive patient audio is processed.

Apian, Synnovis and NVIDIA are teaming up to use advanced computing and digital simulations using NVIDIA Cosmos to help hospital robots handle the heavy lifting of moving lab samples, freeing up medical staff to focus on patient care.

Danaher and NVIDIA are collaborating to advance autonomous science and help accelerate the pace of discovery by bringing accelerated computing, scientific and physical AI into laboratory workflows and instruments. Danaher’s recently announced AI-powered autonomous lab — designed to deliver up to 8x faster discovery of affinity reagents — is using NVIDIA BioNeMo open models within its AI-driven protein design workflows, data and automation technologies that connect design with experimental feedback.

Smart Cities

On the civic side, Liverpool City Region is building an NVIDIA-powered “City Brain” — a sovereign AI strategy designed as a blueprint for dozens of Unitary Authorities across the U.K.

Digital Jersey, Innovation City Belfast and Tees Valley Combined Authority are each exploring how trusted AI compute can serve communities and public services.

Financial Services

In financial services, LSEG and NVIDIA are bringing GPU-accelerated market analytics to global institutions.

Revolut introduced PRAGMA earlier in the year — a foundation model for banking event sequences, pre-trained on 26 million anonymised user histories. PRAGMA has moved from research to real-time scaling.

Lloyds Banking Group built ATUM, a transaction foundation model that converts billions of banking transactions into reusable AI. And Starling is developing a banking foundation model trained on Civo’s U.K. sovereign cloud infrastructure.

NatWest is collaborating with NVIDIA on Nemotron Voice to better understand how banking customers across the U.K. speak, while Modulr is deploying a network of agents using Nemotron for distinct parts of payment operations.

Quantum

Quantum is moving from research to systems deployable alongside classical AI, and the U.K. is at the heart of this shift.

Riverlane has shown how GPUs can accelerate decoder recalibration workflows used in quantum error correction, helping decoder performance keep pace with real-time changes in quantum hardware behaviour. The work forms part of a broader collaboration exploring how AI, accelerated computing and quantum error-correction technologies can work together to support future quantum systems.

Together, these announcements showcase how enterprises, organizations and startups in the U.K. are fostering innovation across the nation. This includes more than 3,100 U.K. companies that are a part of the NVIDIA Inception program for cutting-edge startups.

Tune in to NVIDIA founder and CEO Jensen Huang’s GTC Berlin keynote on Wednesday, Oct. 21, at 11 a.m. CEST.

NVIDIA Announces New Investments in the UK Startup Ecosystem

Rocket Software, a prominent player in the U.S. software industry, has successfully acquired the Israeli startup Geniez AI, in a deal estimated to be worth up to $100 million. This acquisition comes just a year after Geniez AI announced its Seed funding round. The company, which boasts a compact team of only six employees, was co-founded by two alumni from Mamram, the renowned computing unit of the Israel Defense Forces.

In September 2025, Geniez AI secured $6 million during its Seed funding round, co-led by StageOne Ventures and Canapi Ventures. Gil Peleg, the CEO, along with co-founder Dan Shprung, previously established Model9, a company specializing in mainframe data management, which was acquired by BMC Software in 2023.

Geniez AI’s innovative technology focuses on generative AI applications tailored for enterprise-level management and modernization of mainframe systems—crucial computing infrastructure that supports vital operations in various sectors, particularly in financial services. Their AI-driven agents function directly on mainframes, monitoring real-time systems, diagnosing issues, and executing workflows through natural language commands.

Following the acquisition, Geniez AI’s technical leadership and engineering talent will integrate with Rocket Software, helping to expedite the company’s AI initiatives, including enhancements to Rocket EVA, their platform dedicated to agentic AI within core operational systems. The collaboration seeks to transition mainframe management from a reliance on specialized expertise to AI-supported operations that simplify the complexity of these systems for easier management and modernization.

Rocket Software emphasizes that this acquisition aims to tackle the growing shortage of skilled professionals in the mainframe domain while also addressing operational complexities and automating critical workflows. The consolidation of technologies is designed to empower teams to leverage institutional knowledge into actionable insights, facilitating informed decision-making across core systems while ensuring security, governance, and control.

Peleg expressed enthusiasm about the acquisition, underscoring Geniez AI’s foundational belief in the potential of GenAI-based applications to transform enterprise understanding and management of mainframe environments. He noted that partnering with Rocket would provide the necessary scale, customer outreach, and mainframe expertise to deploy their solution across a wider range of organizations, paving the way for the next generation of intelligent, mission-critical operations.

Milan Shetti, the president and CEO of Rocket Software, remarked that this acquisition would significantly boost their capability to securely implement agentic AI into enterprise systems. He characterized the deal as a pivotal part of Rocket’s overarching strategy to modernize core systems while enriching AI functionalities for the essential workloads they support.

Israeli AI startup with six employees, founded a year ago, bought for about $100 million.

A sunlit Tel Aviv startup office: two suited executives shake hands over a polished table next to an oversized mock check reading “$100,000,000.” Behind them, a large “NovaAI” logo is frosted on a glass wall, and a single Israeli flag stands in the corner. Six empty workstations with closed laptops line the side, bathed in warm natural light to mark this one-year-old, six-employee AI startup’s big exit.

One-year-old Israeli AI startup with six employees acquired for estimated $100 million - CTech

- Rocket Software has acquired Geniez AI for up to $100 million to enhance AI initiatives.  
- Geniez AI specializes in generative AI technology for modernizing enterprise mainframe systems.  
- The acquisition aims to address mainframe professional shortages and reduce operational complexities.  

U.S. software company Rocket Software has acquired Israeli startup Geniez AI in a deal valued at up to $100 million, according to estimates. The two founders, both graduates of Mamram, the Israel Defense Forces’ computing unit, received the acquisition offer one year after announcing their Seed round, with the company having a team of only six employees.

In September 2025, Geniez AI raised $6 million in a Seed funding round co-led by StageOne Ventures and Canapi Ventures. CEO Gil Peleg and co-founder Dan Shprung previously founded Model9, a mainframe data management startup acquired by BMC Software in 2023.

1 View gallery Geneiz founders. ( David Garb )

Geniez AI develops generative AI technology designed to help enterprises manage and modernize mainframe systems, the large-scale computers that underpin critical business operations across industries, including financial services. Its AI agents operate directly on mainframes, monitoring live systems, diagnosing incidents and executing workflows using natural language.

As part of the acquisition, Geniez AI’s technical leadership and engineering team will join Rocket Software to help accelerate its AI initiatives, including new capabilities for Rocket EVA, the company’s agentic AI platform for core systems. The companies aim to move mainframe management away from processes that depend heavily on specialized expertise toward AI-assisted operations that make complex systems easier to understand, manage and modernize.

Rocket Software said the acquisition would help enterprises address the shortage of skilled mainframe professionals, reduce operational complexity and automate mission-critical workflows. The combined technology is intended to help teams turn institutional knowledge into actionable intelligence, surface insights and take action across core systems while maintaining appropriate security, governance and control.

“Geniez AI was founded on the belief that GenAI-based business applications can fundamentally change how enterprises understand and operate their mainframe environments,” Peleg said. Joining Rocket, he added, would give the company the scale, customer reach and mainframe expertise needed to bring its technology to more organizations and help shape the next era of intelligent, mission-critical operations.

Milan Shetti, president and CEO of Rocket Software, said the acquisition would accelerate the company’s efforts to help enterprises securely apply agentic AI to their systems. He described the deal as part of Rocket’s broader strategy to modernize core systems while extending AI capabilities to the critical workloads that run on them.

One-year-old Israeli AI startup with six employees acquired for estimated $100 millio

Could the saga of Decart's potential sale be approaching a positive conclusion? Just a month ago, the Israeli AI startup faced a significant setback when Anthropic opted not to proceed with an acquisition, despite the due diligence reportedly being finalized. This decision raised concerns about Decart's technology and future business potential, casting a cloud over the company, which has only been in operation for three years.

Decart attributed the deal's collapse to its unwillingness to shift operations to the U.S., though this explanation was met with skepticism. Following the failed negotiation with Anthropic, Decart sought to reinitiate talks with Nvidia, which had previously expressed interest in acquiring the company before Anthropic's involvement. Unfortunately for Decart, Nvidia declined to revisit negotiations, prompting the company's shareholders to turn their attention to Amazon instead. Recent discussions with Amazon indicate that a potential agreement might be reached by the end of October, valuing Decart at $5.5 billion—a figure below the $7 billion reported by The Wall Street Journal and less than what Anthropic had considered offering.

Even at the lower valuation, this sale would signify a tremendous success for a company of Decart's age, especially one generating millions in revenue through various projects and commercial partnerships while still refining its product strategy. The startup's latest funding round had pegged its worth at $4 billion, signaling that a sale would afford investors a considerable premium and significant returns in a brief timeframe. Notably, Decart's shareholders consist primarily of American venture capital firms, with fewer Israeli backers.

The question arises: why is Amazon keen on acquiring Decart, particularly after Anthropic's last-minute withdrawal? The motivations here might differ from those of Anthropic. Decart has maintained a collaborative relationship with Amazon since its early days and is developing software that can enhance the performance of the tech giant's AI chips.

Amazon aims to diminish its reliance on Nvidia by creating its own Trainium processors, and Decart’s technology is tailored to optimize the performance of AI models across various chip architectures. This innovation enables clients to maximize the efficiency of their existing hardware. Decart was also one of the first to showcase an AI model performing better on Amazon's chips compared to Nvidia’s GPUs, a feat demonstrated by co-founder Dean Leitersdorf during Amazon’s major annual event last year.

In 2015, Amazon made a pivotal acquisition of Israeli chipmaker Annapurna Labs for around $350 million, an investment that played a vital role in building Amazon's in-house silicon capabilities. The relationship between Amazon and Annapurna transformed its custom-chip operations, fostering better control over the hardware in its data centers and minimizing reliance on external suppliers.

Amazon's renewed interest in Decart represents a strategic effort to replicate its earlier success in the AI domain. While Annapurna focused on hardware development, Decart’s niche lies in enhancing the performance of existing AI models across various processor types.

Historically, Amazon has been cautious regarding large acquisitions, favoring smaller transactions over blockbuster deals. A purchase of Decart would signify a substantial endeavor, even at a valuation less than the previously mentioned $7 billion. However, the attraction may hinge less on acquiring a singular product and more on securing a respected team and innovative technology to bolster Amazon's competitive edge in AI computing.

With fewer than 100 employees, Decart's engineers are seen as integral to helping Amazon optimize its chips and manage diverse processor workloads. This capability is expected to gain importance as cloud providers strive to offer alternatives to Nvidia's market dominance.

Decart operates in two primary sectors. Initially, its technology for real-time video generation garnered significant attention. The startup is part of a select group of Israeli firms developing proprietary large AI models. Its breakthrough product, Oasis—a game where AI creates a virtual landscape based on player interaction—helped establish Decart’s reputation and rapidly attracted users, reaching a million downloads within days and surpassing ChatGPT at the time.

Despite its initial success in gaming, Decart sought to pivot toward a broader business model. This led to the emergence of its second focus area: software that enhances AI chip efficiency, reduces costs, and facilitates AI workloads on varied architectures. The startup collaborates closely with chip manufacturers, making its technology valuable for companies looking to challenge Nvidia’s position or minimize dependency on its processors.

This explains Nvidia's past interest in investing in Decart and pursuing an acquisition. The company has also strengthened ties with industry giants like Amazon and Google. For Amazon, the rationale is simple: if its chips can efficiently handle AI tasks, they become more appealing to customers. Software that simplifies the process of switching between chip architectures can enhance its cloud platform's attractiveness.

A deal with Amazon at $5.5 billion would mark an impressive milestone for a startup just three years in. It would deliver considerable returns for early investors, including founders Dean and Orian Leitersdorf and Moshe Shalev. Orian, who joined soon after the startup’s inception while still completing his military duty, and Yossi Sariel—who commanded Unit 8200 before joining Decart—helped propel the company from its early days.

Decart quickly attracted significant venture capital interest, with Sequoia Capital leading a $21 million seed round and further backing from Oren Zeev. Within a month, the startup secured an additional $25 million at a valuation of about $500 million from Benchmark, an influential U.S. venture firm that had previously been less active in Israel.

Shortly after this, Decart's valuation jumped to $3.1 billion, with total funding reaching $153 million. By May 2026, it secured another $300 million, bringing its cumulative funding to over $450 million at a $4 billion valuation.

Amazon could acquire Decart for $5.5 billion following Anthropic's withdrawal.

A bright conference room flooded with natural morning light through floor-to-ceiling windows. On a sleek wooden table lies a stamped “$5.5 B” contract bearing the Decart logo. Behind it, a large Amazon logo hangs on the wall. At the table’s edge, an empty chair with a toppled “Anthropic” nameplate suggests its representative just walked away.

Amazon may get Decart for $5.5 billion after Anthropic walked away - CTech

- Decart's acquisition talks have shifted from Anthropic to Amazon, valuing the startup at $5.5 billion.  
- Amazon seeks to enhance its AI chip efficiency through Decart's innovative technology solutions.  
- A sale would deliver significant returns for investors in the rapidly growing Israeli startup.

Is Decart’s sale saga heading toward a happy ending? A month ago, Israeli AI startup Decart suffered one of the biggest setbacks in its relatively short history when Anthropic decided not to acquire the company. The decision came after due diligence had reportedly been completed, casting a shadow over the three-year-old startup and raising questions about its technology and business prospects.

Decart offered a patriotic explanation for the collapse of the deal, saying it had been unwilling to relocate to the United States. Few, however, found that explanation convincing. Calcalist has learned that after the Anthropic deal fell through, Decart tried to reopen negotiations with Nvidia, which had made the first serious acquisition offer before Anthropic entered the picture.

Nvidia declined to return to the negotiating table, prompting Decart’s shareholders to pursue a new round of talks, this time with Amazon. Those discussions have brought the company closer to a deal that could be signed by the end of October. According to information obtained by Calcalist, the proposed transaction would value Decart at $5.5 billion, below the $7 billion figure reported by The Wall Street Journal and less than Anthropic was reportedly prepared to pay.

1 View gallery Decart founders Orian Leitersdorf (from right), Dean Leitersdorf and Moshe Shalev. ( Amit Elkayam )

Even at $5.5 billion, a sale would represent a remarkable outcome for a three-year-old company generating tens of millions of dollars in revenue, primarily through projects and commercial engagements, while still working to establish its ultimate product strategy and business model. Decart’s most recent funding round valued it at $4 billion, meaning the proposed sale would give investors a substantial premium and potentially allow them to realize significant returns in a remarkably short time. Its shareholders are predominantly U.S. venture capital funds, with relatively few Israeli investors.

Why would Amazon want to acquire Decart, particularly after Anthropic’s decision to abandon a deal at such an advanced stage? A sale to Amazon may make strategic sense for reasons that differ from Anthropic’s potential motivations. Decart has worked with Amazon almost since its inception and is developing technology that could help the tech giant improve the performance of its AI chips.

Amazon is trying to reduce its dependence on Nvidia by developing its own Trainium processors. Decart’s software is designed to make AI models run more efficiently across different chip architectures, potentially allowing customers to extract more performance from the hardware they already have. The startup was among the first to demonstrate an AI model running more efficiently on Amazon’s chips than on Nvidia GPUs. About a year ago, co-founder Dean Leitersdorf demonstrated the integration of Decart’s technology with Amazon’s new chip at the tech giant’s flagship annual event. The two companies have maintained a close relationship since.

In 2015, Amazon acquired Israeli chipmaker Annapurna Labs for approximately $350 million. The deal became one of the most consequential acquisitions in the development of Amazon’s in-house silicon business. The Yokneam-based company helped lay the foundation for Amazon Web Services’ custom-chip operations, enabling the cloud giant to gain greater control over the hardware powering its data centers and reduce its dependence on external suppliers.

Amazon’s reported interest in Decart reflects a potential effort to replicate that success in the AI era. Whereas Annapurna developed chips, Decart focuses on optimizing the performance of AI models running on processors made by different manufacturers.

Amazon has historically been reluctant to spend heavily on acquisitions, preferring smaller deals to mega-transactions. A purchase of Decart would therefore represent a significant commitment, even at a valuation below the $7 billion figure reported in the United States. The appeal may lie less in acquiring a single product than in gaining control of a highly regarded team and technology that could strengthen Amazon’s competitive position in AI computing.

Decart employs fewer than 100 people, but its engineers are seen as capable of helping Amazon get more out of its own chips and manage workloads across different types of processors. That could become increasingly valuable as cloud providers compete to offer customers alternatives to Nvidia’s dominant hardware.

Decart has two main areas of activity. The product that initially attracted the most attention was its technology for generating video in real time. The company is among a relatively small group of Israeli startups developing its own large AI models, and it first gained widespread recognition with Oasis, a game in which AI generates a virtual environment as the player interacts with it.

The demonstration helped put Decart on the map before the company was even a year old. Oasis reached one million users in three days and surpassed ChatGPT in downloads during that period.

The early success generated revenue, but Decart did not intend to become a gaming company. It spent the following years seeking a broader business direction that could turn its underlying technology into a scalable product.

That effort helped bring its second area of activity to the fore: software designed to improve the efficiency of AI chips, reduce computing costs and allow AI workloads to run across different chip architectures. Decart works closely with chip manufacturers, making its technology potentially valuable to companies seeking to compete with Nvidia or reduce their dependence on its processors.

That helps explain Nvidia’s investment in Decart and its earlier interest in acquiring the startup. Decart has also deepened its relationships with Amazon and Google. For Amazon, the attraction is particularly clear: its own chips are more valuable if customers can run AI workloads on them efficiently, and software that makes it easier to move between chip architectures could strengthen the appeal of its cloud platform.

A sale to Amazon for $5.5 billion would be a remarkable outcome for a startup founded only three years ago and would deliver a significant return to investors who have backed it through a rapid rise in valuation. It would, however, be a less dramatic milestone for Israel’s AI industry than a deal with Anthropic might have been. An acquisition by Anthropic, which has not established a major development presence in Israel, could have provided a new entry point for one of the world’s leading AI companies into the local ecosystem. Amazon, by contrast, already has a substantial development operation in Israel.

In an interview with Calcalist after Decart was named the most promising Israeli startup of 2025, founders Dean Leitersdorf and Moshe Shalev described their ambition in sweeping terms: “We want to be the Google, Apple, or Facebook of the AI world.” They spoke of the transformative impact of technologies and institutions that changed the world, and of their desire to build a company with similarly lasting influence.

Amazon already employs an estimated 2,000 people in Israel, with development operations in Tel Aviv and Haifa. If the acquisition goes ahead, Decart is unlikely to be remembered as the company that first put Israel on the AI map. Instead, it would add another major exit to the long list of successful deals generated by the country’s technology industry.

The sale would also deliver a rapid return for Decart’s shareholders, including its founders, Dean and Orian Leitersdorf and Moshe Shalev, as well as investors who backed the company from its earliest days. Orian, Dean’s brother, joined the startup shortly after its founding while still completing his military service. Yossi Sariel, who commanded Unit 8200 on October 7, had previously commanded the founders and later joined the company after the conclusion of his military career.

Decart attracted major venture capital firms almost immediately after its founding. Sequoia Capital led an unusually large $21 million seed round, with participation from Oren Zeev. Just a month later, the company raised another $25 million at a valuation of approximately $500 million. That round was led by Benchmark, a major U.S. venture capital firm that had until then had limited activity in Israel.

Less than a year later, Decart raised additional capital at a valuation of $3.1 billion. That round brought its total funding to $153 million. In May 2026, it raised another $300 million at a $4 billion valuation, bringing total funding to more than $450 million.

Amazon may get Decart for $5.5 billion after Anthropic walked away

A recent report by CNBC's Kate Rooney reveals that both Amazon and Nvidia are in discussions about potential acquisitions of AI startups. This development highlights the increasing interest of major tech companies in bolstering their artificial intelligence capabilities through strategic investments in emerging firms within the sector.

Reports indicate Amazon and Nvidia are exploring AI startup partnerships.

In a sunlit high-rise boardroom, two executive silhouettes lean over a polished glass table. On the table rests a softly glowing AI microchip prototype. Behind them, a giant Amazon logo banner on one wall and an equally prominent Nvidia logo banner on the opposite wall. Warm natural light floods through floor-to-ceiling windows, revealing a bustling city skyline and an air of anticipation for a breakthrough AI startup partnership.

Amazon and Nvidia considering AI startup deals: Reports - CNBC

- Amazon and Nvidia are exploring potential deals with AI startups.  
- Reports indicate interest from both companies in advancing AI technology.  
- CNBC's Kate Rooney provided insights into these potential partnerships.  

In this video

Share Share Article via Facebook Share Article via Twitter Share Article via LinkedIn Share Article via Email

Amazon and Nvidia considering AI startup deals: Reports

CNBC’s Kate Rooney reports that Amazon and Nvidia are eyeing some AI startups.

Nvidia is launching an AI hub in Stockholm for start-ups addressing health and climate issues.

More from AI Startups & Investments

25-year-old Israeli entrepreneur divests stealth AI company for potentially $35 million without revealing details.

Disney Compensates 66 Character.AI Employees with $60 Million Following Tech Job Cuts

Implications of this bankruptcy for AI startups

Ramp achieves $60 billion valuation as it expands further into AI.