AI Daily: Alibaba, Tencent cleared to buy Nvidia’s H200 chips

Catch up on the top artificial intelligence news and commentary by Wall Street analysts on publicly traded companies in the space with this daily recap compiled by The Fly.

Meet Samuel – Your Personal Investing Prophet

NVIDIA CHIPS: Although the U.S. has approved about 10 Chinese companies to purchase Nvidia (NVDA) H200 chips, no deliveries have occurred, leaving the sales in limbo as Jensen Huang joins Donald Trump on a trip to meet Xi Jinping in hopes of breaking the impasse, Reuters reports. Chinese companies including Alibaba (BABA), Tencent (TCEHY), ByteDance, and JD.com (JD) have been cleared to purchase the H200 chips, according to sources.

CISCO BLOWOUT: Cisco Systems (CSCO) is outperforming on Thursday after the networking-equipment company delivered a better-than-anticipated sales forecast and announced plans to cut thousands of jobs while shifting more focus to the fast-growing AI market.

LEGAL FIGHT WITH APPLE: Apple’s (AAPL) two-year-old partnership with OpenAI has become “strained” as the AI startup fails to see the expected benefits from the deal, people familiar with the matter told Bloomberg’s Mark Gurman. OpenAI lawyers are actively working with an outside legal firm on a range of options that could be formally executed in the near future, which could include sending the iPhone maker a notice alleging breach of contract without necessarily filing a full lawsuit to begin, the sources told Gurman.

MYTHOS HELPS FIND BUGS: Security researchers say they have discovered a new way of circumventing Apple’s (AAPL) – state-of-the art security technology, using techniques they discovered while testing an early version of Anthropic’s Mythos AI software in April, The Wall Street Journal’s Robert McMillan reports. The researchers with Calif say the software they wrote links together two bugs and a handful of techniques to corrupt the Mac’s memory and then gain access to parts of the device that should be inaccessible. Apple, which is deploying and testing frontier AI models to test and patch vulnerabilities, is reviewing the Calif report to validate its findings, the author notes. “Security is our top priority, and we take reports of potential vulnerabilities very seriously,” a company spokeswoman said.

GROK CHATBOT: Elon Musk’s xAI has recruited several Wall Street firms to test its Grok chatbot, including Apollo Global (APO) and Morgan Stanley (MS), Carmen Arroyo of Bloomberg reports, citing people familiar with the matter. While banks are signing up for Grok, some people familiar with the matter say financiers are rarely using the chatbot for work. Musk’s xAI is working hard to boost revenue by selling chatbot subscriptions and access to its computing resources, since much of its sales to date come from deals with Musk’s other ventures.

INCEPTION ACQUISITION: Microsoft (MSFT) is exploring acquisitions of AI startups as the company positions for a future less dependent on OpenAI, though concerns over regulatory scrutiny led it to abandon discussions around Cursor due to overlap with GitHub Copilot, Reuters’ Deepa Seetharaman, Milana Vinn, and Kenrick Cai report. Microsoft is in discussions with Inception, a small startup built by a Stanford University team focused on a different method of developing large language models, three people familiar with the matter say. SpaceX also courted Inception, people familiar with the matter say, adding that Inception has hired a bank to help negotiate a deal and is looking for a price of over $1B.

AGENTIC AI OS: Fiserv (FISV) launched agentOS, an agentic AI operating system designed to help financial institutions deploy, manage, and scale AI agents across their banking workflows. Six financial institutions have partnered with Fiserv to co-develop agentOS, with two running agents in beta. agentOS is expected to be widely available by August. AgentOS is supported by strategic collaborations with OpenAI and Amazon (AMZN) Web Services to bring frontier AI to financial institutions at scale.