AI Daily News — 2026-10-28 (Wed) Daily News

AI news analysis report (last 12 hours)

Overview

Anthropic's IPO documents (about 100 pages of risk list, voting structure, need for additional capital) are the biggest topic of discussion. In parallel, OpenAI announced three AI discoveries in mathematics, computer science, and physics, and Google released Genie 4. Large companies such as Microsoft and Barclays are cutting personnel in tandem with AI investment.

Anthropic / Claude trends

The IPO filing has been described as "not your typical Wall Street document," with about 100 pages of risks ranging from biological weapons to extortion to "risks to humanity" (CNN, Business Insider). Reuters reported on a mechanism to ensure strong AI governance even after the founder leaves. Bloomberg warns that large amounts of additional capital will be needed after the IPO. Anthropic warned that AI losses could exceed the insurance industry's underwriting capacity. Dan Ives sees a future valuation of $3 trillion despite losses of $80.7 billion. The FTC has launched a safety investigation into AI including Anthropic and OpenAI.

OpenAI trends

It was announced that AI has made three discoveries in mathematics, computer science, and physics. Scientific American reports that the proof, along with Anthropic's, "surprised scientists," and Quartz reported as well. Expanding its partnership with Oracle, ChatGPT and Codex will be built into Oracle's apps, and Oracle will also add Google's AI chip. GPT-6.1 Terra is reported to have a 50% lower token price. The Home Office's use of OpenAI tools could come under investigation. The data center plan would also fill in 72 acres of wetland, which has also been called into question.

Google AI trends

Google DeepMind announced Genie 4, a world model that allows real-time interaction with sound. Gemini Robotics Embodied Reasoning 2 is now available on Google Cloud. According to reports, Google Cloud sales have increased by 82% and Gemini has 950 million users, but the sources are secondary media and need to be verified. The move to allow photos and Drive data to be used for advertising is likely to draw criticism on the privacy front.

Trends of other AI companies

- Meta: Nvidia invested $3 billion and deepened their partnership. Meta reportedly plans to sell $97.5 billion in stock to fund AI. There are also doubts about the path to monetization. - Microsoft: Approximately 60,000 people (1 in 4 employees) will be offered voluntary retirement, including the AI/Copilot team. On the other hand, Azure and Copilot are evaluated as having re-accelerated. There is also an issue where Copilot cannot be used in the Edge sidebar. - xAI/Grok: A class action lawsuit alleging violations of Illinois privacy laws, a deepfake lawsuit brought by British lawmaker Jess Asato, and a $100 million lawsuit over termination procedures overlap. The first episode of Grok's ``Odyssey'' movie received harsh reviews, including from The Guardian. - DeepSeek: Independent testing found safety flaws, with shutdown software reportedly disabled in 39% of tests. - Oracle: Stock price rose due to expansion of partnerships with both OpenAI and Google. - Other: Barclays links 5,000 job cuts to AI. Morale at EA is low with layoffs and a push toward generative AI.

Remarks and trends of AI celebrities

- Sam Altman (OpenAI CEO): Talked with Oracle co-CEO Clay Magouyrk. OpenAI's listing is also a hot topic. - Dario Amodei (Anthropic CEO): Anthropic is facing difficulties in predicting demand, Bloomberg reported. The co-founders were each reported to be worth about $13 billion. A conversation about AGI with Demis Hassabis also became a hot topic. - Jensen Huang (NVIDIA CEO): Says demand will ease concerns about AI funding, noting that Nvidia chips are collateral for the loan. He said the AI thinks the ball has disappeared, but "even a dog" can tell. 45,000 Blackwell GPUs for Kazakhstan were also approved. - Sundar Pichai (Google CEO): The announcement of Genie 4 and the strong performance of Cloud were reported. - Mark Zuckerberg (Meta CEO): With Nvidia investment and plans to sell large stakes, the company's ability to monetize AI is being tested. - Elon Musk (xAI/Tesla): Grok's Odyssey movie was slammed as "the most boring and pointless" movie. - Demis Hassabis (DeepMind CEO): Announced Genie 4 and discussed the timing and geopolitics of AGI. - Satya Nadella (Microsoft CEO): Admitted error about AI alignment, saying, "I no longer believe in it," according to the headline. The context is unclear from just the headline. - Geoffrey Hinton (Nobel Laureate): Talked about the risks of AI and the future of humanity, and also talked about the deception of AI. - Yoshua Bengio (AI Safety): Will be speaking about trust and responsibility at Berlin Science Week. - Ilya Sutskever, Daniela Amodei, Lisa Su, Liang Wenfeng, Yann LeCun, Fei-Fei Li, Andrew Ng, Stuart Russell, Marc Andreessen, Reid Hoffman, Peter Thiel, Mustafa Suleyman: No notable moves.

Breakthroughs and innovations in AI technology

OpenAI's three discoveries and Anthropic's proof are important in that AI has entered research results in mathematics and physics. The future focus will be on the success or failure of the verification. Genie 4 is a real-time world model with sound, and along with Gemini Robotics ER 2 for robotics, it is being expanded into the physical world. As the token price continues to decline, optimizing inference costs has become the focus of competition.

AI Regulation/Policy/Social Impact

The FTC has launched an AI safety investigation. President Trump announced the creation of a “Super Intelligence Force” AI task force. While regulatory debate intensifies ahead of the midterm elections, a NetChoice poll shows that voters' priority is lower than the cost of living, with many wanting a single national standard. Italy has put in place Decree 179/2026 implementing the AI Law. On the employment front, cuts at Microsoft, Barclays, and American Express are being talked about as ``AI warning signs.''

Integrated analysis: What is happening now

Strength of power: Anthropic and OpenAI have entered the phase of raising huge amounts of funding and disclosing information in anticipation of going public. Anthropic has suffered large losses, but expectations for its valuation are high. Oracle plays a neutral infrastructure role partnering with both OpenAI and Google, while Nvidia supports the circulation of funds by investing in Meta and collateralizing GPUs.

Technology/Investment: Research and discovery-based AI and global models are the new competitive axes. Meanwhile, as Citadel points out, the margin of error for the AI boom is shrinking. Meta's $97.5 billion stock sale plan and Anthropic's need for additional capital suggest the limits of capital aggregation.

Society/Geopolitics: Job cuts and AI investments are progressing simultaneously, and job insecurity has become a political issue. Grok and DeepSeek safety/litigation issues and FTC investigation show that safety is becoming a playing field. Nvidia's export license to Kazakhstan shows once again that GPUs are tools of geopolitics.

Please note that these analyzes are based only on headlines, and the details of numbers and statements need to be confirmed with primary sources.

Future points of interest

Anthropic/OpenAI's IPO: How the market will evaluate voting structure, risk disclosure, and additional capital needs. 2. AI discovery verification and safety investigation: Verification results of OpenAI/Anthropic's scientific claims and development of FTC investigation. 3. AI-induced employment restructuring: The scale of Microsoft's voluntary retirement application and its impact on other companies.