AI Daily News — 2026-06-24 (Wed)

# AI industry trend report (last 12 hours) *June 24, 2026*

Overview

Anthropic announces Claude Tag, a Slack integration tool for businesses, marking a new phase in AI in the workplace. On the other hand, the major players were divided, with DeepSeek emerging as China's top-rated AI startup and xAI/Grok facing financial damage to SpaceX. Google is facing the double whammy of a loss of top talent and a decline in its online dominance, and a seismic shift in the battle for AI supremacy is underway.

Anthropic / Claude trends

Claude Tag is the main news. It is a tool that functions as a "virtual employee" within the Slack workspace, providing work support while continuously learning internal conversations and documents. TechCrunch described it as "learning a company's Slack messages one by one," and cited its deep integration into the company's culture. At the same time, a large-scale failure occurred in Claude's main model, causing a temporary suspension of service, but it has since been restored. Anthropic also participated in the Cannes Advertising Festival, demonstrating the penetration of AI into the brand marketing field. Another notable talent acquisition is the transfer of Nobel Prize winner John Jumper (co-inventor of AlphaFold) from DeepMind to Anthropic.

OpenAI trends

As an example of the medical application of GPT-5, OpenAI officially announced a case study in which immunologist Derya Unutmaz solved a three-year-old unresolved problem using GPT-5. OpenAI is reportedly in talks with the White House about a new model called Mythos, but authorities are taking a wait-and-see approach. We are also expanding our efforts in the cybersecurity field, and have started a program that allows partner companies to incorporate GPT-5.5. Amazon's decision to cancel the release of the original film ``Sam Altman biopic'' out of consideration for its partnership with OpenAI has sparked debate over the media dominance of AI Big Tech.

Google AI trends

CNBC reported that ``cracks are appearing in Google's dominance in the AI era,'' and the erosion of Google's position in the search market began to appear in numbers. DeepMind continues to lose top talent, and Fortune magazine has questioned its ability to maintain its competitiveness as a research institution. However, CEO Demis Hassabis countered that ``Google is still acquiring top AI talent.'' On the product side, Gemini continues to expand its ecosystem, with Gemini adding automatic fixes for broken formulas in Google Sheets, Nokia announcing a partnership to embed Gemini-based AI agents into network management products, and TCL equipping European TVs with Gemini.

Trends of other AI companies

Meta: Zuckerberg continues his wearable strategy with the introduction of $299 low-cost Ray-Ban AI smart glasses. Meanwhile, investors are expressing skepticism about the more than Rp 248 billion AI investment. The company also announced the launch of a prediction market app.

xAI/Grok: Multiple media outlets report that xAI's financial losses are spreading to SpaceX's parent company. A series of articles with the tone "Grok undermines SpaceX" have brought to light the problem of resource allocation within Elon Musk's empire.

DeepSeek: Reportedly, it has emerged as the highest-rated AI startup in China. The Harness team is rushing into the AI agent race. Geopolitical concerns continue, with the city of Ghent in Belgium banning the use of Grok and DeepSeek for its employees.

Microsoft: Announces the next phase of cloud operations with the theme of Agentic Observability. There are also reports that AI adoption is outpacing security response capabilities.

Remarks and trends of AI celebrities

Jensen Huang (NVIDIA CEO): Attended a Senate AI hearing and said, ``The era of Agentic has begun, and it's an incredible time for software companies.'' NVIDIA's positive news for Microsoft (AI breakthroughs other than chips) was also reported.

Demis Hassabis (Google DeepMind CEO): ``AGI may arrive around 2030'' (ZME Science). Responding to the talent leak report, ``Google is still winning in talent acquisition'' (Semafor). There was no direct comment on the leak of Nobel Prize winner John Jumper.

Mark Zuckerberg (Meta CEO): Led the launch of $299 smart glasses and demonstrated ambition for the Meta prediction market app.

Elon Musk (xAI Founder): SpaceX reportedly plans to borrow $20 billion after $86 billion IPO. SoftBank CEO questioned, ``Is there any point in having a data center in space?'' Speculation about a Tesla-SpaceX merger also surfaced.

Geoffrey Hinton (AI Godfather/Nobel Laureate): ``Regulations are a steering wheel, not a brake,'' said (UCD lecture), asserting the need for regulations to guide AI.

Dario Amodei (Anthropic CEO): As a related article, a commentary analyzing the possibility of the upcoming AI IPO bursting the bubble was published. The issue of the AI employment crisis is also being raised.

Sundar Pichai (Google/Alphabet CEO): His name was linked to a CNBC report about Google search's declining dominance in the age of AI. There are also reports that Google AI Overviews presents fiction as fact.

Sam Altman (OpenAI CEO): Amazon's cancellation of the release of a biopic has led to criticism that ``billionaires are controlling culture.''

Breakthroughs and innovations in AI technology

NVIDIA has announced an "AI breakthrough" in fields other than chips, and it has been pointed out that it could be a catalyst for demand for next-generation data centers (the specific details are unknown, but it appears to be a software/platform layer other than semiconductors). At BIO 2026, NVIDIA announced the BioNeMo toolkit (drug discovery support for AI agents). The Linux Foundation has announced its intention to launch "Agent Name Service," a reliable ID platform for AI agents, and infrastructure development for the multi-agent era is progressing. Google DeepMind has released a video called ``When millions of AI agents meet,'' hinting at large-scale multi-agent research. Analysis of whether GLM-5.2 can cause a new "DeepSeek moment" is also attracting attention.

AI Regulation/Policy/Social Impact

New York State is asking companies to prepare for three AI laws affecting the workplace. Rhode Island passes an opt-out law for AI scribes for medical records. The EPA is moving to delegate regulatory authority for AI data centers to states and local communities, creating a regulatory vacuum at the federal level. Privacy issues (WSJ) and the failure of strategy without AI governance (Law.com) are also widely discussed. In terms of the environment, it was pointed out that the AI revolution's ``water, electricity, and waste'' problems are becoming more serious on an unprecedented scale. Geoffrey Hinton's statement that "regulation = steering wheel" clearly expresses the mainstream position of the AI safety community.

Integrated analysis: What is happening now

Changes in the balance of power: Google's long-standing technological superiority has begun to be questioned due to the combination of reports that Google's ``dominance will collapse in the AI era'' and the exodus of personnel from DeepMind (such as John Jumper). Anthropic, on the other hand, is stepping up its offensive with a strategy of deep integration into corporate infrastructure called Claude Tag and recruiting talent from outside (Jumper).

A new phase of corporate AI: As epitomized by Claude Tag's Slack integration, AI is moving from being an external chatbot service to becoming a "permanent member of the company." Continuous access to corporate data and accumulation of context will become a source of competitive advantage.

The rise of Chinese AI: DeepSeek's rising valuation, the arrival of GLM-5.2, and reports that Microsoft is considering adopting DeepSeek internally continue to shake the assumption of Western AI dominance. While Belgian municipalities are moving to ban DeepSeek due to geopolitical concerns, cost advantages are likely to encourage its actual adoption.

xAI/Musk's predicament: Reports that Grok's financial losses are spilling over to SpaceX's parent company demonstrate the heavy cost of AI development and the difficulty of commercialization. Combined with increased borrowings after the SpaceX IPO, Musk's empire's cash flow management is now being questioned.

Acceleration of Agentic Shift: Multiple players, including NVIDIA's Jensen Huang, Linux Foundation's Agent Name Service, and Microsoft's agentic observability, are simultaneously working to "prepare the foundation for the agent era." The second half of 2026 is likely to be the year when commercial deployment of AI agents will begin in earnest.

Future points of interest

Google DeepMind's chain of talent leaks: How far will the "competition to recruit talent from Google", symbolized by John Jumper's transfer to Anthropic, accelerate in the future? The sustainability of the Hassabis system is being questioned.

Claude Tag and Copilot's battle for corporate AI supremacy: Anthropic (Slack) vs. Microsoft (Teams/Copilot) in a battle to integrate corporate infrastructure. Recruitment trends over the past few months will be an important indicator because companies that are deeply rooted in a company's workflow will have long-term stickiness.

OpenAI's "Mythos" model and White House trends: Whether or not approval and partnership progress at the government level will determine the speed at which OpenAI enters the government and defense markets. It will be interesting to see when regulators' silence turns into clear policy.

AI industry trend report (June 24, 2026)

Overview

The biggest focus in the AI industry over the past 12 hours has been the dispute between Anthropic and the National Security Agency over access to sensitive systems. Anthropic's model is leading to a legal dispute over government contracts, with authorities alleging that it was able to sneak into sensitive NSA systems. At the same time, OpenAI released the cybersecurity specialized model GPT‑5.5‑Cyber, and the possibility of collaboration with Microsoft/DeepSeek has also emerged, intensifying the battle for AI supremacy on three fronts: technology, regulation, and security.

Anthropic / Claude trends

The most notable news this time is the conflict between Anthropic and the US government. The NSA lost access to a new Anthropic model called Mythos, with multiple media outlets reporting that "Anthropic's AI penetrated sensitive systems within hours" (Euronews). As a result, Anthropic customers sued the US government over losing access to Fable AI (Bloomberg), and AI startups also filed lawsuits against the federal government (Business Insider). This series of events paradoxically proves the high capabilities of Anthropic's model, with South China media pointing out that ``US restrictions on Anthropic are the best advertisement for Chinese AI'' (South China Morning Post). Dario Amodei reiterated his reason for leaving OpenAI, saying, ``There was no need to argue with someone I don't trust.'' (AOL.com)

OpenAI trends

OpenAI releases cybersecurity specialized model GPT‑5.5‑Cyber. It is said to have a fully autonomous agent function that fully automates everything from vulnerability detection to patch application (CyberSecurityNews). Florida sues OpenAI and Sam Altman in unprecedented lawsuit over AI safety (MSN). On the other hand, legal and personal scandals overlapped, including allegations of conflicts of interest regarding Altman's investment relationship (Crypto Briefing), and director Luca Guadagnino's Sam Altman biopic being dropped by the studio after Amazon withdrew (AOL.com/Tech Times). Amazon's decision to partner with OpenAI is said to be behind the movie's withdrawal, and the fallout from the platform battle is also spreading to the cultural industry.

Google AI trends

Google Cloud announced the integration of Nokia and Gemini AI agents to power autonomous network management (The Fast Mode). In the financial sector, HSBC and Google Cloud have signed a partnership agreement, with returns expected to be up to $100 million per use case (TNGlobal). In the entertainment field, the company invested $75 million in film studio A24, launching the first attempt to incorporate DeepMind's AI tools into film production (LAmag/Tech Times). NKE's development of an AI shopping function aimed at meeting demand for the FIFA World Cup was also well-received (Stocktwits). Sundar Pichai spoke at Stanford's graduation ceremony, causing an unusual situation in which hundreds of students walked out in protest (MSN). In India, he has spoken out about the democratization of AI in discussions with Prime Minister Modi (News On AIR).

Trends of other AI companies

Meta: Announcing new smart glasses products. Equipped with real-time translation in 20 languages and blink correction function (Gulf News). Collaboration with Essilor Luxottica also hints at lowering the price range (Inside Retail Asia). Under pressure from the Trump administration to voluntarily participate in a government security review of the model (Crypto Briefing). In copyright litigation, we are also fighting against the author's prompt appeal application (Law360).

Microsoft: The company is said to be considering the possibility of integrating DeepSeek into Copilot, but discussions continue due to security concerns (BankInfoSecurity). Specifically, we are considering adopting the open source version of DeepSeek V4 (Open Source For You).

DeepSeek: A complete ban in the US is expected to be avoided for the time being (Mashable). Tencent is testing DeepSeek-powered AI agent for WeChat enterprise app (Bloomberg). Remarks by the successor to Chinese Premier Li Keqiang, who appealed to "avoid loss of control" in AI governance, also attracted attention (HKFP).

xAI / Tesla: Tesla has applied for a trademark for Megapod, which could be used as a data center play, and the market is paying attention to it as a rival to NVIDIA (Yahoo Finance). Elon Musk launches additional funding campaign after raising $85 billion (Jerusalem Post).

Remarks and trends of AI celebrities

Sam Altman (OpenAI CEO): A situation where both personal and corporate scandals overlap, including withdrawal from a biopic, a personal lawsuit filed by the state of Florida, and allegations of conflict of interest in investments. There are concerns about the impact on IPO plans (The Edge Malaysia).

Dario Amodei (Anthropic CEO): Clarified the reason for leaving OpenAI as ``the pointlessness of arguing with someone you don't trust.'' Anthropic has made clear its position on maintaining its independence amid tensions with the government.

Jensen Huang (NVIDIA CEO): Remarks evaluating China's role in AI development (GuruFocus). On the other hand, despite the embargo, it has been reported that NVIDIA's banned chips are being traded in the Chinese market at twice the market price, raising questions about the effectiveness of export controls (Startup Fortune). Rising demand for AI chips boosted Wall Street (MSN).

Elon Musk (xAI/Tesla CEO): Megapod trademark application suggests challenge to NVIDIA. Cory Doctorow criticized Musk and other tech entrepreneurs in The Guardian, saying, ``To make billions of dollars, you have to hurt people.''

Sundar Pichai (Google/Alphabet CEO): Students protest by leaving Stanford graduation speech. This event symbolized social divisions over AI ethics and the relationship between universities and industry.

Mark Zuckerberg (Meta CEO): Regarding AI glasses, he said, ``It is necessary for them to function as facewear before they become technology,'' emphasizing a product strategy that prioritizes design (Business Insider).

Demis Hassabis (Google DeepMind CEO): In addition to announcing investment in A24, will attend next week's G7 summit with executives from Anthropic and OpenAI (Business Post).

Yann LeCun (Meta Chief AI Scientist): Pointed out that AI applications in the field of robotics are accelerating towards the era of physical AI. Tech entrepreneurs say they are using the world as a testing ground for robotization (ABC News).

Geoffrey Hinton (AI Godfather): Discussion on the limitations of current LLM based on next token prediction from a linguistic/biological perspective (The India Forum).

Breakthroughs and innovations in AI technology

OpenAI's GPT-5.5-Cyber is equipped with functions that fully automate everything from vulnerability detection to patch application in the cybersecurity field, and is attracting attention as a model that shows the transition from the conventional assisted type to a fully autonomous type. In the field of physical AI, the development from language models to robots with bodies is accelerating, led by Yann LeCun. Dan Ives views this ``AI revolution as the bottom of the third inning'' and emphasizes that the temporary drop in chips is nothing more than a ``gut check'' (Benzinga). If the report that Anthropic's AI "infiltrated" a classified system "in a matter of hours" is true, it would send shockwaves both inside and outside the industry as an indication that the capabilities of AI agents may have reached a practical level for security purposes.

AI Regulation/Policy/Social Impact

The NSA-Anthropic dispute provides unprecedented visibility into the supply chain risks of government agencies' use of cutting-edge AI. Florida's OpenAI lawsuit could set a precedent for state-level legal sanctions against AI companies. Greece is drafting a national framework to implement the EU AI Act (Digital Watch Observatory). In the medical field, the question has arisen: ``Do doctors bear legal responsibility even if AI makes a decision?'' (Medscape) AI regulations in financial services are also being developed (Orrick). The Chinese leader has emphasized that AI governance should avoid "loss of control," and both the United States and China are simultaneously advancing the institutionalization of AI safety management. The Trump administration has asked Meta to voluntarily participate in government safety reviews of AI models, and industrial policy is moving in the gray area between voluntary and forced.

Integrated analysis: What is happening now

The biggest fault line in the AI industry this season is the trade-off between security and access. The incident in which Anthropic's AI allegedly infiltrated the NSA's classified systems shows that the capabilities of cutting-edge models may exceed the agency's expectations. The government wants to use the model but fears uncontrollable risks, and the conflict has manifested itself in contract disputes and lawsuits.

In terms of power dynamics, DeepSeek's avoidance of the ban and the rise of Chinese AI players (with NVIDIA's embargoed chips trading at twice the market price) cast doubt on the effectiveness of the US export control strategy. As South Chinese media points out, there is a paradox in that the more the US tightens restrictions on Anthropic, the greater the relative advantage of Chinese AI will be.

Cyber security and physical AI are rapidly emerging as technology trends. GPT‑5.5‑Cyber indicates that agent-based AI has entered the practical stage in specific specialized areas, and physical AI, promoted by YannLeCun, is becoming the next main battlefield.

On the investment side, AI chip stocks have recovered after a temporary plunge, and long-term bulls, such as Dan Ives' "bottom of the third inning" theory, remain firm. However, as the decline in Meta stock shows, there remains skepticism about when the massive investment in infrastructure will pay off. Geopolitical adjustments are about to enter a new phase, with major AI executives scheduled to attend the G7 summit.

Future points of interest

Anthropic vs. NSA legal dispute — could impact the entire security framework for government AI procurement and redefine the government contracting model for other AI companies.

Remarks by major AI executives at the G7 Summit — As a place where the top executives of Anthropic, OpenAI, and DeepMind gather together, it will serve as a touchstone for building international AI governance consensus.

Microsoft's decision to integrate DeepSeek V4 with Copilot — Whether or not major Western platforms will adopt the open source model from China will be the most stressful decision among technology, security, and diplomacy.