Nvidia has announced a partnership with AI startup Ineffable Intelligence, a company founded by former Google DeepMind reinforcement learning leader David Silver to pursue next-generation AI systems based on reinforcement learning.

The partnership will focus on building infrastructure for large-scale reinforcement learning systems that learn through experience and trial-and-error rather than relying primarily on human-generated data. Nvidia said engineers from both companies will collaborate directly on infrastructure and training pipelines optimized for reinforcement learning workloads.

The London-based startup was founded in late 2025 and emerged publicly earlier this year with a record $1.1 billion seed funding round co-led by Sequoia and Lightspeed. Investors included Nvidia, DST Global, Index Ventures, Google, and the U.K.’s Sovereign AI Fund.

Nvidia said the collaboration will use its Grace Blackwell chips alongside the company’s upcoming Vera Rubin platform to support large-scale training environments.

“The next frontier of AI is superlearners — systems that learn continuously from experience,” Nvidia CEO Jensen Huang said in a statement.

Silver said current AI systems have largely mastered learning from existing human knowledge but still struggle to independently discover new knowledge and strategies through experience.

“Researchers have largely solved the easier problem of AI: how to build systems that know all the things humans already know,” Silver said. “But now we need to solve the harder problem of AI: how to build systems that discover new knowledge for themselves.”

According to the company, Ineffable’s systems may require new model architectures and training methods beyond conventional large language model approaches.

The companies said they will specifically focus on creating scalable pipelines capable of continuously feeding reinforcement learning systems with simulated experiences and environmental feedback.

Reinforcement Learning Returns to Center Stage

Reinforcement learning previously played a central role in major AI breakthroughs including DeepMind’s AlphaGo systems, but recent generative AI advances have been driven largely by large language models trained on massive datasets of human-created text and images.

Several leading researchers now argue that future AI progress may increasingly depend on systems capable of independently exploring environments, testing strategies, and learning from outcomes rather than relying solely on static datasets.

That transition requires significantly different infrastructure, training pipelines, and computational architectures compared with conventional generative AI systems.

New AI Labs Intensify Competition For Talent And Capital

Ineffable is part of a broader wave of new AI research companies launched by former researchers from major AI labs including OpenAI, DeepMind, Anthropic, Meta, and xAI.

Investors have poured billions into these startups over the past year as competition intensifies around superintelligence research, autonomous AI systems, and advanced reasoning models. On the same day as Nvidia’s announcement, AI startup Recursive Superintelligence, founded by former DeepMind researcher Tim Rocktäschel, announced a $650 million funding round.

The surge in funding reflects growing investor belief that the next phase of AI competition may depend not only on larger models, but also on entirely new learning methods, infrastructure systems, and training paradigms capable of producing more autonomous forms of intelligence.

Sam Altman testified Tuesday that Elon Musk abandoned OpenAI during a crucial period in the company’s development, rejecting Musk’s claims that OpenAI improperly transformed itself away from its nonprofit mission.

Speaking for roughly four hours in federal court in Oakland, California, Altman told jurors that Musk failed to follow through on commitments to support the company financially as OpenAI struggled to secure the computing resources needed to compete in artificial intelligence research.

“We were kind of left for dead,” Altman testified.

Musk sued OpenAI, Altman, and OpenAI President Greg Brockman in 2024, alleging the company violated its founding principles by shifting toward commercial operations and pursuing profits rather than operating solely for charitable purposes. Musk argues that the roughly $38 million he contributed to OpenAI was used for unauthorized commercial expansion.

Altman disputed that claim in court, saying he never promised Musk that OpenAI would permanently maintain a nonprofit-only structure.

Much of the trial has focused on internal negotiations in 2017 and 2018 involving Altman, Musk, Brockman, and co-founder Ilya Sutskever over how to finance increasingly expensive AI development. According to testimony, OpenAI leaders debated several possible corporate structures, including for-profit models, as they sought billions of dollars in computing and infrastructure funding.

Those discussions ultimately collapsed, and Musk left OpenAI’s board in February 2018.

Altman testified that Musk’s departure created uncertainty inside OpenAI, with employees worrying about how the organization would survive financially. He also said some researchers viewed Musk’s exit as a morale improvement due to dissatisfaction with his management style.

“I don’t think Mr. Musk understood how to run a good research lab,” Altman told the court.

Court filings and testimony also revealed that Musk continued corresponding with OpenAI leadership after leaving the board. In one 2018 email presented during testimony, Musk wrote that OpenAI had “0%” chance of competing with Google DeepMind without dramatically increasing resources and spending billions annually.

Altman said the message remained “burned into my memory.”

OpenAI’s Origins Face Unprecedented Scrutiny

The trial has become one of the most consequential legal disputes in the AI industry because it directly examines how OpenAI evolved from a nonprofit research lab into one of the world’s most valuable private technology companies.

Musk argues OpenAI abandoned its original public-interest mission in favor of commercial expansion tied closely to Microsoft and large-scale investor funding. OpenAI has countered that evolving its structure was necessary to finance advanced AI development and compete against heavily funded rivals.

Testimony from current and former executives has exposed years of internal disagreements over governance, funding, safety priorities, and control of increasingly powerful AI systems.

Earlier in the trial, Sutskever testified that he had previously gathered evidence alleging Altman showed a “consistent pattern of lying” before Altman’s temporary removal as CEO in 2023. The court also heard testimony about previously undisclosed discussions involving a potential merger between OpenAI and rival Anthropic after Altman’s brief ouster.

AI Governance and Corporate Control Move into Public View

Beyond the personal conflict between Musk and Altman, the case highlights broader tensions shaping the AI industry as companies balance nonprofit ideals, investor demands, infrastructure costs, and control over frontier AI systems.

The enormous computing requirements associated with advanced AI development have pushed leading labs toward increasingly commercial models and deep partnerships with cloud providers and investors. At the same time, regulators, policymakers, and courts are beginning to examine how these organizations govern technologies that could have major economic and national security implications.

The outcome of the trial could influence how future AI companies structure governance, investor oversight, and nonprofit commitments as the industry continues consolidating around a small group of heavily capitalized firms.

Anthropic has overtaken OpenAI in verified business customer adoption for the first time, according to new data from fintech platform Ramp.

Ramp’s latest AI Index, which analyzes expense data from more than 50,000 companies using its payment and finance platform, found that 34.4% of participating businesses are now paying for Anthropic services, compared with 32.3% for OpenAI. It marks the first time Anthropic has held the top position in the survey.

According to Ramp economist Ara Kharazian, Anthropic had already established a lead among highly technical industries including finance, technology, and professional services before expanding into broader enterprise categories.

The data also illustrates how rapidly Anthropic has grown over the past year. In May 2025, only 9% of surveyed businesses were paying for Anthropic products. That figure has since increased by 26 percentage points over a 12-month period. During the same timeframe, OpenAI’s business adoption share declined by roughly 1%, while overall enterprise adoption of AI products across the survey increased by 9%.

Kharazian said Anthropic’s strategy of initially focusing on technical customers and developer-oriented use cases helped establish stronger traction within enterprise environments before broader expansion through products such as Claude Cowork.

The findings align with broader market signals suggesting growing enterprise adoption of Claude models. OpenRouter usage rankings, which track another segment of AI users and developers, last showed OpenAI ahead of Anthropic in December 2025.

Ramp noted that the index is not a complete representation of the overall AI market because it only reflects companies using its platform. However, the dataset remains one of the largest publicly discussed indicators of verified commercial AI spending activity.

Enterprise AI Competition Shifts Toward Deployment And Reliability

The report highlights how competition between leading AI companies is increasingly being shaped by enterprise deployment rather than consumer visibility alone.

Anthropic has spent much of the past year expanding its presence in regulated industries and enterprise workflows, particularly in finance, cybersecurity, operations, and software development. Its Claude family of models has gained traction among businesses seeking longer context handling, coding assistance, and AI agents designed for workplace tasks.

The company has also aggressively expanded enterprise infrastructure and partnerships in recent months, including new deployment initiatives, financial services AI agents, cybersecurity tools, and large-scale compute agreements aimed at supporting business demand.

Meanwhile, OpenAI continues to maintain a dominant consumer footprint through ChatGPT while simultaneously pushing deeper into enterprise deployments through consulting partnerships, deployment services, and productivity integrations.

Amazon has introduced Alexa for Shopping, a new AI-powered shopping assistant designed to combine conversational AI, product expertise, and customer shopping history into a unified retail experience across Amazon’s app, website, and Echo Show devices.

The launch merges capabilities from Alexa+ and Amazon’s Rufus shopping assistant, which the company said helped more than 300 million customers research and compare products in 2025. Alexa for Shopping is now integrated directly into Amazon’s main search bar, allowing users to ask conversational questions, compare products, track orders, generate shopping guides, and automate purchases using natural language.

Amazon said the assistant continuously personalizes recommendations using browsing activity, purchase history, preferences, and conversations across Alexa-enabled devices. The company described the system as a persistent shopping layer that carries context between devices and sessions instead of resetting interactions each time a customer searches.

The assistant can create AI-generated category overviews, compare products side-by-side from search results, surface one-year price history charts, and automatically monitor products for price drops. Customers can also create “Scheduled Actions” that automate recurring shopping tasks such as replenishing household items, tracking book releases, or adding products to carts when prices reach specific targets.

Amazon is also expanding agentic shopping capabilities through Shop Direct and its “Buy for Me” feature. The system can discover products from external retailers and, for eligible items, complete purchases automatically using stored payment and shipping information.

The company said Alexa for Shopping can also generate personalized shopping guides for complex purchases such as laptops, TVs, or appliances by summarizing reviews, features, pricing differences, and category insights across Amazon and the broader web.

In addition to mobile and desktop support, Amazon is bringing the full Amazon storefront experience to Echo Show devices for the first time. Customers can browse and purchase products using voice commands, touch controls, or a combination of both.

Alexa for Shopping is rolling out to all U.S. customers this week and does not require a Prime membership, Alexa app subscription, or Echo device.

Amazon Pushes AI Deeper Into Commerce Automation

The launch marks one of Amazon’s most aggressive attempts yet to transform e-commerce from search-based navigation into AI-assisted decision making and task automation.

Rather than relying on keyword searches and static filters, Alexa for Shopping is designed to function as a persistent shopping assistant that remembers preferences, previous conversations, recurring purchases, family information, and shopping behavior across Amazon’s ecosystem.

Features such as automated cart-building, conversational product research, and price-triggered purchases move Amazon closer to agentic commerce systems where AI actively manages portions of the shopping process on behalf of users.

The integration of Rufus product intelligence with Alexa+ personalization also gives Amazon a broader contextual data advantage across retail, smart home devices, and media services.