AI News Archive: June 26, 2026 — Part 4
Sourced from 500+ daily AI sources, scored by relevance.
- The Government Boot Is Coming Down on AI
"We don’t believe this kind of government access process should become the long-term default," OpenAI said in a Friday announcement.
- China’s Zhipu AI sparks new ‘DeepSeek moment’ with cost-effective coding model
Nearly a year and a half after China’s DeepSeek shook Silicon Valley with its powerful yet affordable artificial intelligence model, Beijing-based Zhipu AI has delivered another jolt to the US tech industry. American entrepreneurs and researchers are praising the coding performance and cost-effectiveness of Zhipu’s new flagship model, GLM-5.2. Released earlier this month, the model’s release is being hailed by some as a new “DeepSeek moment”, with users calling it the first-ever open-weight...
- Living without an AI kill switch
Living without an AI kill switch The Japan Times
Score: 60🌐 MovesJun 26, 2026https://www.japantimes.co.jp/commentary/2026/06/26/world/an-ai-kill-switch/ - Anthropic Thinks Its Own Success Is Key to Making AI Safe
Anthropic's critics argue it's rapidly accumulating power. The company says that's what responsible AI development looks like.
Score: 60🌐 MovesJun 26, 2026https://www.wired.com/story/anthropic-thinks-ai-can-only-be-safe-under-its-control/ - Next‑Gen ADAS/AD Architectures: Power, Networking, Safety & Sensors
Join this webinar and learn how high‑performance semiconductor technologies support centralized sensor fusion and reliable ADAS systems. The post Next‑Gen ADAS/AD Architectures: Power, Networking, Safety & Sensors appeared first on EE Times .
Score: 60🌐 MovesJun 26, 2026https://www.eetimes.com/next%e2%80%91gen-adas-ad-architectures-power-networking-safety-sensors/ - Oracle cut workforce by 21,000 over past year amid AI spending push
Oracle cut workforce by 21,000 over past year amid AI spending push Austin American-Statesman
Score: 60🌐 MovesJun 26, 2026https://www.statesman.com/business/technology/article/austin-oracle-ai-layoffs-2026-22317147.php - Italy probes AI-fueled price hikes in Microsoft 365
Regulator says subscribers may have been defaulted onto more expensive plans with Copilot features attached
Score: 60🌐 MovesJun 26, 2026https://www.theregister.com/saas/2026/06/26/italy-probes-ai-fueled-price-hikes-in-microsoft-365/5262986 - Google Cloud UK summit showcases agentic AI initiatives across different industries
Google Cloud UK summit showcases agentic AI initiatives across different industries verdict.co.uk
- S Korea firm Semifive wins $8m AI chip order from Japan
The order is roughly 10% of Semifive’s 111.8 billion won (US$72.7 million) in consolidated revenue for 2024, based on the company’s disclosed results.
Score: 60🌐 MovesJun 26, 2026https://www.techinasia.com/nvidia-rival-rebellions-taps-jpmorgan-south-korea-ipo - SAP aligns commerce data for AI personalisation
SAP aligns fragmented commerce data structures to enable operational AI personalisation at the execution layer. Enterprise leadership routinely establishes objectives to anticipate customer requirements and deliver relevant interactions across digital touchpoints. However, the actual infrastructure running inside these enterprises fails to support systematic execution at the required volume. Recommendation engines display generic product listings because […] The post SAP aligns commerce data for AI personalisation appeared first on AI News .
Score: 60🌐 MovesJun 26, 2026https://www.artificialintelligence-news.com/news/sap-aligns-commerce-data-for-ai-personalisation/ - Zhipu AI vs MiniMax: China's Anthropic and OpenAI Mirror the Valuation Reversal
Zhipu AI and MiniMax see their market values diverge as the two Chinese AI giants increasingly mirror the Anthropic vs OpenAI dynamics playing out globally
- Cathay Cargo Expands Capacity to Meet AI Hardware Demand
Cathay Cargo Expands Capacity to Meet AI Hardware Demand Caixin Global
Score: 60🌐 MovesJun 26, 2026https://www.caixinglobal.com/2026-06-27/cathay-cargo-expands-capacity-to-meet-ai-hardware-demand-102458151.html - ‘Developers are moving from writing code to orchestrating AI agents’: AWS’s Jeff Barr
‘Developers are moving from writing code to orchestrating AI agents’: AWS’s Jeff Barr
- AI-driven cyber threats reshapes enterprise cybersecurity spending in India
As enterprises adopt AI at scale, cybersecurity spending is shifting from compliance and perimeter defence to identity protection, AI governance, cloud security, and continuous threat monitoring
- How Telangana police is using air-gapped AI systems for child sexual abuse investigations
Telangana Cyber Security Bureau deploys C-SIGHT, an offline AI platform developed with Vatins Systems, to investigate child sexual exploitation material with 98% accuracy in under 20 minutes. The post How Telangana police is using air-gapped AI systems for child sexual abuse investigations appeared first on MEDIANAMA .
- OpenAI hasn't held pre-IPO investor meetings or set timeline yet, sources say
OpenAI said it confidentially filed its prospectus with the SEC earlier this month, but noted it "may be a while" before it goes public.
- How the Iran war impacts Abu Dhabi's AI strategy
Abu Dhabi's ambitions to turn the UAE into a global hub for digital infrastructure and AI, dubbed "UAE AI Strategy 2031," face pressure after the war with Iran. But the UAE is also known for its business resilience.
Score: 60🌐 MovesJun 26, 2026https://www.dw.com/en/how-the-iran-war-impacts-abu-dhabi-s-ai-strategy/a-77722632?maca=en-rss-en-all-1573-rdf - Amazon just made a key AI cloud service more expensive
Amazon just made a key AI cloud service more expensive Business Insider
Score: 60🌐 MovesJun 26, 2026https://www.businessinsider.com/amazon-raises-ai-cloud-prices-memory-chip-costs-soar-2026-6 - Robotaxis drive miles just to get cleaned and charged; this new startup wants to fix that
Aseon Labs, which came out of Y Combinator's 2026 spring cohort, has raised $10 million from Crane Venture Partners and others.
- Presight Announces Cohort II of Global AI Companies Joining its AI Accelerator Programme
Presight launches second cohort of AI startups in its accelerator program.
- Deploy a Production-Ready NVIDIA AI-Q Blueprint on Oracle Cloud Infrastructure
AI agents have changed a lot in the last two years. The first could only answer one question at a time. Then came multi-turn chat, where the model could keep...
- OpenAI poaches Uber India chief to lead its biggest market outside the US
The hire marks OpenAI's latest push into India, expanding offices, partnerships and hiring.
Score: 59🌐 MovesJun 26, 2026https://techcrunch.com/2026/06/26/openai-poaches-uber-india-chief-to-lead-its-biggest-market-outside-the-u-s/ - This new research challenges nearly every big AI narrative of 2026
This new research challenges nearly every big AI narrative of 2026 Business Insider
Score: 58🌐 MovesJun 26, 2026https://www.businessinsider.com/enterprise-ai-spending-grows-openai-leads-rbc-reveals-2026-6 - A New Generation Studies AI, Apple's Recipe for On-Device Models, GLM5.2 Tackles Open-Ended Problems
A New Generation Studies AI, Apple's Recipe for On-Device Models, GLM5.2 Tackles Open-Ended Problems
- Should Idaho Police Use AI to Write Reports From Body Cam Footage?
Some police departments in Idaho have started using an AI tool to summarize body camera footage, but legal experts worry that incorrect reports could hurt people’s cases and encourage administrators to cut corners.
- How AI Is Transforming DevOps and Cloud Engineering
How AI Is Transforming DevOps and Cloud Engineering DevOps.com
- India and the GCC corridor: The next growth engine for AI-led insurance transformation
By Dhirendra Singh, Head of Global Operations, Zinnia The conversation around Artificial Intelligence in financial services has moved decisively beyond experimentation. Across the global insurance industry, the focus has shifted […] The post India and the GCC corridor: The next growth engine for AI-led insurance transformation appeared first on Express Computer .
- GOSIM Paris: This Is What Open Source AI Looks Like in 2026
My notes from a day packed with open source AI, inference engines, and a few unexpected takeaways Image from ©GOSIM GOSIM (Global Open Source Innovation Meetup) is an open-source conference whose latest edition took place in Paris on May 5–6 at Station F. It brought together the people who actually build the open-source tools we all rely on. It’s one of those conferences that doesn’t get much attention, but probably deserves more. A few takeaways from a day spent exploring open-source AI. Talk 1: Can an AI Mathematician Be More Than Just a Black Box? by Sir Timothy Gowers LLMs are getting good at math, fast. I was curious to hear what a Fields Medal mathematician actually thought about it. Sir Gowers made a point: a correct proof isn’t enough. He mentioned PhD students feeling overwhelmed by how quickly AI solves problems they’ve spent months on (and the illustration was insane lol), and then an LLM solves it in a matter of minutes. His group at Cambridge is actually working on making AI reasoning transparent and teachable, not just fast. The shift from solve it to show me how you got there feels like the more interesting race right now. Sir Timothy Gowers Talk 2: Data Science Is Not AI, but It Is Its Genesis, and Its New Frontier by Gaël Varoquaux We were in the middle of a conf dominated by agentic AI, and then Gaël Varoquaux (Chief Science Officer at Probabl) walked on stage and made a simple point: you don’t need an LLM for everything. Data science is reinventing itself, and statistics still matter. He introduced TabICL, an open source foundation model for tabular data that holds its own against the best. Probabl, the team behind scikit-learn, is proving that “old school” open source data science is evolving without losing what made it work. In a room full of agent frameworks, that felt like a necessary reality check. Gaël Varoquaux We kicked things off with these two keynotes, which I found particularly relevant. But I’m not going to list every talk one by one here, you can already find the slides online. What I really wanted to do instead was share my thoughts on the conference as a whole. I’ve become a bit of a Paris tech conference enthusiast, attending events that range from deeply technical to much broader ones (I’ll let you guess which are my favourites). vLLM Workshop After this dose of big picture, it was time for the workshops, and I had a plan. Right after the two keynotes, I headed to the technical workshops. I had already planned my day around one session: vLLM. For a bit of context, it’s a library we’ve been using at work for inference for a while now, and the idea behind it is still genuinely fascinating (I’m currently writing an article on LLM inference and fine-tuning, so stay tuned. Yes, this is a shameless plug 🫣). During the workshop, Daniele Trifiro, vLLM contributor and Principal Software Engineer at Red Hat, started by presenting the core idea behind vLLM: PagedAttention, a KV-cache memory management approach inspired by OS paging, designed to drastically reduce memory waste and unlock fast inference. He covered recent optimizations, and I was genuinely surprised by how many new features I hadn’t heard of, and especially by the number of parameters I didn’t know existed, despite having used vLLM regularly. The most interesting part was definitely the live code walkthrough , yes, literally. In big conferences (Not talking about conf like ICML, NeurIPS and so on), workshops are often theoretical, short, and don’t go very deep. This was the complete opposite: we had time to explore the library, and to dive into the hardware side, including vLLM’s support for Google TPU, AWS Neuron, and Intel Gaudi (not just NVIDIA GPUs). He also covered evaluation, a crucial topic in this space, using lm-evaluation-harness to check that vLLM actually meets requirements. The talk was a good opportunity to discuss the community, the importance of open-source contributors, and leaderboards like OpenLLM. I think you get the picture, I wasn’t disappointed. It was exactly the kind of technical deep dive I was hoping for: over an hour and a half of genuinely interesting content that captured the full scope of work behind vLLM. What About Agentic AI? If vLLM is about running a model efficiently, FlagOS tackles a more upstream question: how to run it everywhere. But before getting into that, a quick word on the agentic AI track in general. What would a 2026 conference be without agentic AI? There was something for everyone, from workshops to panel discussions. If you’re starting to know me (or not), panels are not really my thing, and this time I sometimes felt a gap between the advertised topic and what was actually being discussed. You’d walk into a room expecting a conversation about agentic AI and end up hearing about strict European AI regulations. Interesting topic, of course, just not what you came for. On the workshop side FlagOS tackles the challenge of running models across multiple hardware targets with a unified stack. The BAAI team, led by Yonghua Lin (Vice President & Chief Engineer at BAAI), explained how they built the framework, with FlagGems, FlagTree (multi-target compiler), and FlagScale (a non-invasive plugin for frameworks). What struck me most was the circularity of FlagOS. The loop is real: FlagOS runs the agents, and the agents build FlagOS in return. Concretely, complex infrastructure tasks, like 13-step model backporting with end-to-end precision validation, or kernel generation via MCP, are now handled autonomously by agents, with 99% correctness on multiple operator benchmarks. The infrastructure isn’t just agent-ready anymore, it’s agent-built and continuously improved by agents. Fascinating! They then shifted the focus to evaluation via FlagEval and PanEval, a rethought approach to benchmarking (contributed to the Eclipse Foundation). The project is already available here , though still fairly recent. Back to Data Science A complete change of pace here, stepping away from infra and back to the fundamentals of data science. One session that caught my attention in the Probabl track focused on the Skore library. I was already loosely familiar with it, but this talk made me look at it differently. Skore is built for enterprise data science: not the solo notebook, but teams, organizations, and scale. The idea is to make tracking, exploring, and sharing workflows as simple as a fit() / predict(). The session itself was short but dense. What stood out wasn’t the discovery of the library, but the way it was framed. I already knew the basics, but Marie Sacksick, Product Engineer, and Fabien Pesquerel, Developer Relations Engineer at Probabl, surfaced features I hadn’t really paid attention to before, especially around collaboration and workflow tracking. I left with a very concrete idea: turning it into a team presentation during one of our tech watch sessions at work. In just 20 minutes of live coding, it was enough to make me want to dive back into the docs, and potentially contribute. For context, we regularly run open-source Saturdays, often working on Probabl’s Scrub library. Last weekend, I even found myself going through the Skore docs again. That’s usually the sign a conference really did its job. Women in AI & Responsible Open Source To close the day on a high note, something a little different, and a topic I care deeply about. As a volunteer with the Paris chapter of Women in Machine Learning & Data Science (WiMLDS), I was particularly happy to see these topics included in such a technical conference. We spend a lot of time talking about models, benchmarks, and infrastructure, but not nearly enough time talking about the people building them and the communities around them. The discussion, led by my WiMLDS colleague Marie Sacksick and Joanna Kramer, Co-Founder at WISE (Women in Safety & Ethics), revolved around two main themes: Diversity in open source: how to make communities more inclusive and welcoming for underrepresented profiles Responsible AI: open-source tools and approaches for AI fairness and model security What I appreciated most is that the conversation stayed practical. Rather than stopping at broad principles, it encouraged everyone to think of one concrete action they could take within their own community to move things forward. Seeing this kind of discussion alongside deep technical workshops felt important. Open source is ultimately about people as much as it is about code. And that’s a wrap. To sum it up, GOSIM 2026 was one of my favourite conferences this year. It’s highly technical, low on commercial fluff, and driven by a genuine commitment to advancing open source. I loved the diversity of topics, the deeply technical themes, and the formats, with workshops long enough to actually go somewhere. Looking back, a few things stood out. Infrastructure is becoming a real pillar of the AI stack, agentic AI is clearly everywhere, and it was refreshing to see data science still very much part of the picture alongside all the LLM hype. Thank you for reading, and feel free to share your thoughts in the comments! 😊 GOSIM Paris: This Is What Open Source AI Looks Like in 2026 was originally published in Towards AI on Medium, where people are continuing the conversation by highlighting and responding to this story.
- AI Reshapes Corporate Structures and Hollows Out Management, Report Says
AI Reshapes Corporate Structures and Hollows Out Management, Report Says Caixin Global
- Google Finance App for Android Announced With AI Portfolio Tools, Real-Time Market Updates
Google on Thursday announced the Google Finance app for Android. It allows users to track watchlists, monitor real-time market data, and receive AI-generated insights. The company will also expand its capabilities to bring features like live earnings calls and today’s new portfolio in the coming months. Alongside, the web experience has been upgraded with new portfo...
- Backed by Lakestar, Seedcamp and EWOR, SE3 unveils spatial AI platform for autonomous systems
Spatial AI company SE3 Labs today emerges from stealth. SE3 builds foundational spatial intelligence technology for the next generation of autonomous systems. As autonomous systems operate in increas...
- How the DeepMind mafia brought the AI boom to London
The tech sector is buzzing in Britain. But can it ever be more than a US outpost?
- AI, data and trust driving MSME financing: Dhaval Kulkarni, CTO, RXIL
India’s MSME sector is increasingly becoming the backbone of the country’s digital credit ecosystem. As financing volumes grow and supply chains become more interconnected, platforms facilitating MSME financing are evolving […] The post AI, data and trust driving MSME financing: Dhaval Kulkarni, CTO, RXIL appeared first on Express Computer .
- Decoding Trust in Enterprise Language AI. In conversation with Slator
Explores five key aspects of trust in enterprise language AI: accuracy, consistency, security, scale, and outcomes.
- AI and Engineering Jobs: New Research Suggests Apocalypse Is Still Far Away
Contrary to what the world has been articulating over AI replacing software developers, new research suggests a different story could be brewing now. Analysis of more than 80 million companies across the globe suggests that these engineering positions may actually be the most resilient job function. Of course, if that sounds incredulous and makes you […] The post AI and Engineering Jobs: New Research Suggests Apocalypse Is Still Far Away appeared first on CXOToday.com .
- ‘Botsitting’: The AI time-savings killer only governance can stop
One of AI’s biggest selling points is all the high-value tasks employees will be free to accomplish with the time saved using AI. Reality, however, remains far from that. While IT workers and other employees do save several hours each week thanks to AI, more than half of that time is burned up babysitting the technology, a new study reveals. According to a survey from the Work AI Institute , digital workers save an average of 11 hours a week through AI, but the net time savings is much less, because they spend 6.4 hours a week “botsitting.” Botsitting involves activities such as feeding AI tools missing context, checking AI outputs, debugging AI mistakes , rerunning prompts, and cleaning up the confident-but-wrong answers they leave behind, as defined by the Work AI Institute, a research group founded by AI copilot and search provider Glean. The botsitting problem is real, several IT leaders agree, and it has serious implications for IT organizations. In many cases, organizations aren’t training their employees to effectively use AI, says Tal Carmi , CIO at digital adoption platform provider WalkMe. WalkMe’s 2026 State of Digital Adoption report found similar results, with employees losing nearly eight hours a week to botsitting, Carmi notes. At the same time, most employees use AI for shallow tasks like writing emails because they don’t trust it for more complex activities, WalkMe found. As a result, enterprises aren’t getting the full ROI of their AI purchases, Carmi says, a significant issue for CIOs and organizations in general. Hours wasted Going into the survey, researchers at the Work AI Institute suspected botsitting was a problem for many organizations, but the results were eye-opening, according to Rebecca Hinds , founder of the organization. “The surprise was how prevalent it is,” she says. “The fact is that workers are spending roughly the same share of their AI time botsitting as they are using the technology to move work forward.” Moreover, while 87% of digital workers, and 97% of IT workers, said they use AI at their jobs, only 13% believe their use of the tools has led to significantly improved performance or outcomes. Part of the problem is a phenomenon Hinds calls “coordination neglect.” Employees often focus on their own productivity without considering the broader benefits to the organization, she says. As a result, their AI-assisted work sometimes conflicts with another employee’s work. “I can use the technology to, say, convert a single bullet point into a five-page report,” she says. “I can then ship that five-page report to a colleague, but the colleague sees that it’s so much content. They can use the same AI tool to then convert the five-page report back into a series of bullet points.” In some cases, employees do divert AI time savings to personal activities, but the most common use of the time saved, according to survey respondents, is to improve the quality of their work, Hinds says. Overall, however, organizations aren’t seeing a major quality improvement, she adds. Shipping AI-generated work that workers haven’t verified, don’t fully understand, or can’t confidently stand behind is a significant issue, according to the AI Work Institute report. And then there’s the “AI toggle tax” — when employees switch between multiple AI tools to do their jobs, which leads to additional unverified work. Moreover, as employees become overwhelmed with AI tool sprawl , they cognitively offload their work to AI. “They hand more of their thinking and judgment over to the machine,” the report says. “They start to cut corners. They stop checking outputs, verifying sources, and asking whether the AI’s recommendations make any sense.” Governance problems at the core Botsitting, and giving in to AI slop, are real but also symptoms of a larger governance problem, says Frank Meltke , CEO of digital transformation consulting firm contraco. “Workers are spending nearly a full day verifying AI output because nobody at deployment defined what verification was required, who owned it, or what good output looks like before it moves downstream,” he says. “That is a governance gap, not a tool problem.” Meltke also doubts there’s a net time-savings gain of four-plus hours per employee each week when their fellow workers sometimes must redo their AI-assisted outputs. More than two-thirds of digital workers surveyed admit to shipping AI-assisted outputs they have not verified, he notes. “That output lands on someone else downstream, usually without context to fix it,” he says. “The 4.6-hour net gain at the individual level gets absorbed invisibly at the team level as rework nobody budgeted for.” This phenomenon explains why time savings observed by individual employes does not show up in organizational performance, he adds. “The productivity gain was never real savings,” Meltke says. “It was a transfer of labor from the person who generated the output to the person who inherited it.” Not all botsitting is a bad thing, however, says Adam Wachtel , CTO at HR platform Click Boarding. Verifying outputs, iterating on prompts, and adding domain context for the AI tool to use are good engineering practices, when done right, he notes. “The issue is that organizations aren’t distinguishing between what’s worth doing versus a symptom of a poorly deployed tool,” he says. A big problem is a lack of context for AI tools, he suggests. “When AI tools don’t have access to accurate data and aren’t built in the right way to make their output usable, employees become the integration layer that re-explains a project to every tool and fixes what breaks,” Wachtel says. Meanwhile, the 6.4 hours spent botsitting aren’t evenly distributed and instead fall on employees already engaged in detailed work, such as senior engineers, he says. “You have others skipping that verification, thinking they’re saving 11 hours, and then may not be responsible for the mess that comes of it — often downstream when code breaks or a process stops working,” he adds. Individual productivity gains don’t automatically add up to organizational ones, Wachtel adds. For example, if an engineer builds code faster, someone else may have to verify it. One employee’s time savings creates work for someone else. Many organizations also struggle to measure quality of AI outputs, he adds. IT leaders should educate the full C-suite on the metrics that matter the most, rather than how many times an AI tool was used , he recommends. “Organizations are touting efficiency gains, but I don’t see a lot of chatter around agents’ accuracy, continuous improvement, or cost takeout that are more impactful to align to,” he says. “A lot of AI was developed and launched for speed rather than for impact, and so the right people weren’t involved or trained.”
Score: 56🌐 MovesJun 26, 2026https://www.cio.com/article/4188575/botsitting-the-ai-time-savings-killer-only-governance-can-stop.html - Tesla and Waymo duel in the robotaxi race — but the company spending the most builds no cars at all
Uber is quietly writing $500 million checks to lock in robotaxis as Waymo threatens to leave it behind.
- What Is AI Distillation and Why Is It a Worry for the Industry?
Artificial intelligence companies in the US have spent hundreds of billions of dollars to develop more advanced chatbots, betting they can earn enough from customers to justify the investment. That approach carries the risk of being undermined by rivals who build competing AI systems for far less.
- Ford had to rehire 350 engineers after its AI got vehicle quality wrong
Ford has admitted that it had to rehire experienced engineers after its AI systems failed to deliver the quality the company expected. Charles Poon, Ford’s VP of vehicle hardware engineering, told reporters that the automaker mistakenly believed it could swap in AI and still produce a high-quality product. The admission, first reported by The Verge, […] This story continues at The Next Web
Score: 56🌐 MovesJun 26, 2026https://thenextweb.com/news/ford-rehired-350-engineers-ai-quality-jd-power - MagicShot.ai Highlights Growing Demand for AI-Powered Real Estate Marketing Tools as Sellers Navigate Low-Engagement Zillow Listings
MagicShot.ai Highlights Growing Demand for AI-Powered Real Estate Marketing Tools as Sellers Navigate Low-Engagement Zillow Listings USA Today
- Over 57% of MSMEs view AI as a Vital Scale Driver; 25% Already Deploy AI Tools
India’s mammoth MSME sector, which contributes over a third of the country’s GDP, has emerged as a key driver of AI adoption, a new survey reveals. More than 57% of small and medium enterprises surveyed view AI as a core tool for enhancing and accelerating business growth while 25% of them have integrated it into […] The post Over 57% of MSMEs view AI as a Vital Scale Driver; 25% Already Deploy AI Tools appeared first on CXOToday.com .
- We Built the Hardest Test in Human History to Measure AI. It Lasted 18 Months.
Every time researchers built a benchmark to measure how intelligent AI had become, AI broke it. So they built a harder one. Then AI broke… Continue reading on Towards AI »
- Epic Games CEO criticises AI disclosure label on Steam, calls Valve “irresponsible”
Epic Games CEO Tim Sweeney has criticised Valve's Steam AI disclosure policy, calling it a "Scarlet Letter" that stigmatises developers and puts smaller studios at a fatal disadvantage in an increasingly costly market. The post Epic Games CEO criticises AI disclosure label on Steam, calls Valve “irresponsible” appeared first on MEDIANAMA .
- How to win at AI (if you’re not the US or China), with AI minister Kanishka Narayan
Specialisation and research can give the UK leverage
- How Antonio Neri turned HPE into an unlikely AI stock
The veteran dealmaker says he’s ‘building the best networking business on the planet’.
Score: 55🌐 MovesJun 26, 2026https://www.semafor.com/article/06/25/2026/how-antonio-neri-turned-hpe-into-an-unlikely-ai-stock - AI Is Flooding Security Teams With Findings—That Doesn’t Mean They’re Safer
Faster does not always mean safer, and finding more vulnerabilities is not the same thing as reducing meaningful exposure.
- Gate Launches Gate.AI Full-Lifecycle Large Model Management Platform: Strengthening Unified Access and Enterprise Governance
Gate Launches Gate.AI Full-Lifecycle Large Model Management Platform: Strengthening Unified Access and Enterprise Governance USA Today
- Meta Put Kylie Jenner’s Voice in Its AI Glasses. It’s a Smart Lesson in Customer Acquisition
The built-in distribution is way more than a celebrity endorsement.
- You know OpenAI and Nvidia. These are the AI companies building everything else
OpenAI, Anthropic, Google DeepMind, and Nvidia have largely defined the AI narrative for much of the last four years, and for good reason. Nvidia’s chips power virtually every major AI training run in the world. OpenAI’s ChatGPT became the fastest consumer application in history to reach 100 million users. Anthropic has turned Claude into one of enterprise AI’s most trusted model families. Google DeepMind’s AlphaFold solved a protein-folding problem that stumped scientists for 50 years, work that earned CEO Demis Hassabis a Nobel Prize in Chemistry in 2024. But those companies are not the whole story. A broader ecosystem is building the infrastructure, governance layers, and application platforms that make AI commercially viable. These companies do not dominate the news cycle. But the industry increasingly depends on them. When DeepSeek rattled markets last year, the lesson went beyond model efficiency. It was a reminder that serious AI work is no longer confined to a handful of famous companies and research labs. What follows is a look at some of the companies building the rest of the AI stack. NOT MADE IN AMERICA Silicon Valley has long treated frontier AI as a homegrown invention. Mistral AI, building one of the world’s fastest-growing AI businesses from Paris, is challenging that assumption. Founded in April 2023 by Arthur Mensch, formerly of Google DeepMind, and Guillaume Lample and Timothée Lacroix, both formerly of Meta, Mistral built its early reputation on a provocative claim: Frontier-quality models did not require the compute budgets of American hyperscalers. Its open-weight models, released for anyone to build on, helped prove the point. Revenue grew from roughly $10 million in 2023 to more than $400 million in annualized recurring revenue by early 2026. In September 2025, Mistral closed a €1.7 billion Series C led by ASML, pushing its valuation to €11.7 billion. It has since launched Vibe , an agentic platform for research, drafting, and code deployment, and is exploring its own chip design. “Physical infrastructure control matters just as much as underlying model quality for capturing long-term value,” Mensch says. Tokyo-based Sakana AI is pulling the frontier further still. Founded in 2023 by David Ha, who led Google Brain’s research team in Japan, and Llion Jones, one of the co-authors of the influential research paper Attention Is All You Need , Sakana was built around the idea that AI systems could evolve the way nature does, through collective intelligence and constant adaptation rather than raw scale. In March 2025, its AI Scientist system became the first AI to have a paper it independently wrote accepted by peer reviewers at a premier machine learning conference. In November, it raised $135 million at a $2.65 billion valuation, backed by MUFG, Lux Capital, and In-Q-Tel, the CIA’s venture arm. Ha has argued that “innovation, based on constraints”—not unlimited compute—is where the real opportunity lies. Earlier this week, Sakana moved Fugu and Fugu Ultra out of beta and into general availability. The system routes tasks across a pool of frontier AI models instead of training one from scratch, all through a single interface. Sakana’s case for that approach leans on a recent warning sign: On June 12, the U.S. Commerce Department issued an export control directive that forced Anthropic to shut down Claude Fable 5 and Claude Mythos 5, just three days after launch. As of this writing, neither has been restored. “Relying on a single company’s APIs for critical infrastructure, finance, or governance is a material vulnerability,” the company said in its launch announcement . “This risk is no longer a hypothetical possibility, but a reality.” Sakana’s own benchmarks compare Fugu Ultra against Fable 5 and Mythos Preview, using whichever of the two scores higher on a given test. Neither model is part of Fugu’s agent pool, since neither is publicly accessible. The comparison excludes Mythos 5, the more capable model suspended alongside Fable 5, with independent analysis showing it outperforming Fugu Ultra on several of the same benchmarks. THE UNGLAMOROUS INFRASTRUCTURE LAYER What powers enterprise AI at scale is not only the frontier model at the top of the stack. It is also the infrastructure beneath it, the layer that determines whether AI applications can access, reason over, and act on data in ways that produce real business outcomes. For most enterprises, that problem starts before the first line of AI code is written. Organizations adding AI to existing systems are often told to move their data out of relational databases, the traditional table-based systems most business software still runs on, and into new infrastructure. In practice, that can mean multi-year projects. At the pace AI is moving, says RavenDB CEO Oren Eini, that timeline is already a liability. “At the current pace of AI, a two-year project is already too late,” he says. “By the time you ship, the rest of the market has already moved on.” With RavenDB, Eini says AI projects that previously stretched past six months are now being completed in weeks. The drag, he argues, is usually not the AI model. It is the infrastructure around it. Most business databases were never built to let AI search and reason over their data. Even databases that offer a vector index, which lets AI search by meaning instead of exact keywords, only solve part of the problem. “If your database only provides a vector index, all of that burden falls on you,” he explains. RavenDB, founded in 2009, has spent the last two years embedding vector search, AI agents, and generative AI capabilities directly into the database layer. The point is to let businesses build on top of what they already run instead of replacing it. “You shouldn’t have to rebuild your systems to use AI,” Eini says. Alation is solving a related problem one layer up. The data intelligence platform counts Truist Bank, Sallie Mae, Cisco, and Daimler Truck North America among its more than 500 customers. Its work starts from a basic reality: Many enterprise AI projects fail at the data layer. Gartner estimates that 85% of AI projects fail due to poor data quality or lack of relevant data. “An agent without business context is just a clever chatbot,” says CEO Satyen Sangani. “It can answer questions. It can’t drive an outcome that someone in finance, operations, or compliance will sign off on.” The failure mode Sangani keeps seeing has a name: Pilot purgatory. S&P Global found 42% of companies abandoned most of their AI initiatives in 2025, up from 17% the year before. The culprit, consistently, is that organizations started with the technology and then went looking for problems to solve with it. The ones breaking through did the opposite. “Every agent action, every human correction, every data product becomes knowledge that makes the next decision better,” Sangani says. “The companies that get this right become indispensable, because their value grows the longer customers use them.” THE ACCOUNTABILITY LAYER Enterprise AI adoption is moving faster than the governance structures designed to contain it. Between 2023 and 2024, the amount of corporate data uploaded or pasted into AI tools rose by 485%. Gartner projects that over 40% of AI-related data breaches by 2027 will stem from unapproved or improper generative AI use. Several companies are working at the edge of that accountability problem. Two in particular, operating in different domains, are converging around the same issue: AI sprawl, or what happens when enterprises deploy AI faster than they can govern it. Reco, an Israeli agent security and AI governance company, was built to solve that problem. Its platform maps the full AI agent footprint across enterprise SaaS environments, showing what is running, what it can access, and how it is behaving. “Across our customer base, organizations typically discover roughly ten times more apps and a hundred times more agents than they expected when they first connect,” CEO Ofer Klein tells Fast Company. The security leaders getting this right have stopped asking “what AI tools are my employees using?” and started asking a harder question: When an agent has simultaneous access to Salesforce data, SharePoint files, and Slack channels, how do you stop a breach before it starts? The answer starts with knowing where AI actually lives inside the enterprise. “Autonomy without governance is exposure at scale,” Klein says. “Every major platform your enterprise already runs on is now also an agent platform. That is exactly where we show value.” Terzo is making a version of the same case in a different part of the enterprise stack: the contracts, invoices, and purchase orders governing how large organizations spend money. According to World Commerce and Contracting, companies lose an average of 9.2% of annual revenue to poor contract management every year, not through fraud, but through missed rebates, lapsed pricing schedules, and supplier agreements nobody was monitoring. Terzo’s platform connects contracts, invoices, purchase obligations, and ERP data, validating financial activity against contractual terms in real time. According to CEO Brandon Card, one Fortune 50 company found more than $100 million in savings after running its supplier relationships through Terzo’s platform. As AI agents increasingly handle procurement and financial workflows, they are only as trustworthy as the data they operate on. An AI agent working from inaccurate or unstructured contract data does not just make mistakes. It makes them at scale. Terzo is building the financial truth layer meant to prevent that. “The companies that win in the next decade will not just have AI models,” Card says. “They will own the financial intelligence layer that allows autonomous agents to move trillions of dollars safely, accurately, and automatically across the global economy.” BEYOND TEXT For all the progress AI has made in the last few years, almost all of it happened in text—reading it, generating it, and searching through it. However, while text is structured and easy to feed into a model, voice and video are neither. They’re messy, happen in real time, and until recently, almost none of it was kept around long enough to be useful. That gap has been enormous, and mostly invisible. Billions of customer service calls are placed every year, and most are logged and forgotten. Hundreds of millions of surveillance cameras run worldwide, and most of what they capture is never watched. The information was there. The tools to use it were not. That’s changing now. Two companies are closing that gap—one in conversation, and the other in physical space. PolyAI, founded in London by Nikola Mrkšić, a Cambridge machine learning PhD and first engineer at VocalIQ, builds enterprise voice AI for organizations managing large volumes of customer interactions. The company counts more than 100 enterprise customers and runs more than 2,000 live deployments across 45 languages in over 25 countries. A Forrester study found its customers achieve 391% ROI with an average $10.3 million in savings. Deloitte recognized it as the U.K.’s fastest-growing AI company in 2025. In December 2025, it raised $86 million in a Series D with Nvidia’s venture arm among the participants. “PolyAI started with a simple idea: Enterprises should sound human,” Mrkšić said at the announcement. “We turned that idea into reality, and it led to something far greater: The emergence of the agentic enterprise.” He has said publicly that within five years, 90% of contact center work will be automated, a claim that would have sounded reckless three years ago and now sounds like a timeline. Lumana is doing something similar in the physical world. Its AI platform transforms existing security cameras, including those in retail stores, healthcare facilities, and factories, into real-time intelligence systems without new hardware. “Instead of recording footage and hoping someone reviews it later, the system understands what it sees—the behaviors, context, and anomalies—and does so in real time,” says company CEO Sagi Ben Moshe. “A camera isn’t just capturing video anymore. It’s generating intelligence.” In under 18 months, the California-based company scaled to more than 50,000 cameras and says its platform can reduce non-actionable physical access control alarms by as much as 95%. More recently, it has expanded beyond detection and into agentic AI, with specialized agents that can monitor environments, investigate incidents, verify events, and initiate responses without requiring a human operator to continuously review video feeds. Adam Scraba, Nvidia’s head of physical AI ecosystem and vision agents, describes Lumana as turning existing infrastructure into “a perceptive, highly scalable virtual workforce of video analytics AI agents that can understand, verify, and summarize what’s happening in the physical world.” That means cameras stop being a passive cost center, and become instead an active intelligence asset, informing decisions about customer behavior, staffing, and workplace safety. At scale, that could make the physical world as legible to AI as the digital one. THE REAL AI ECONOMY The story of AI value creation that most people tell goes like this: Powerful models emerge, developers build on top of them, enterprises adopt, and economic value follows. What that story leaves out is the middle or context layer: the infrastructure, governance, security, and operational intelligence that makes enterprise adoption viable at scale. McKinsey estimates AI could add up to $4.4 trillion annually to the global economy through enterprise productivity gains alone. But those gains do not materialize just because a company signed an OpenAI contract. They materialize when the data feeding those models is governed and trusted, when the agents operating inside enterprise systems are visible and accountable, when contracts are operationalized rather than stored, and when voice and physical environments are converted into actionable intelligence. That is the work the companies in this piece are doing. In previous technology cycles , the companies that won over the long term were not always the ones with the most famous products. They were the ones those products could not run without. If history is any guide, the AI cycle will not be different.
- AI Budget Increases, Fable’s Potential, AI’s Cyber Frontier: Takeaways From Big Technology’s AI Summit
Don't expect the spending to slow down anytime soon, and more lessons from a day with the leaders of today's biggest AI companies.