President Donald Trump announced Friday that the US government may take equity stakes in AI giants such as OpenAI, Anthropic, and xAI, with Anthropic reportedly absent from these equity talks despite being valued at nearly ₹8 lakh crore. According to CNBC, Trump plans to host AI executives at the White House to discuss the ownership plan as early as next week, framing the idea as a way to give taxpayers direct exposure to AI profits. The proposal comes as both Anthropic and ChatGPT-maker OpenAI race to sell shares on the stock market, with OpenAI CEO Sam Altman having discussed the concept with administration officials since early 2025. The administration has precedent for this approach, having taken roughly 10% stakes in Intel in 2025 and holding positions in IBM and several quantum firms.
In a significant development, Anthropic has proposed that the world's top artificial intelligence companies coordinate a pause in advanced AI system development if the technology becomes too dangerous. As reported by Anthropic, the company's internal research institute plans to explore this issue in collaboration with others and 'take actions' to help build the systems for a credible slowdown or pause. The proposal comes as Anthropic and ChatGPT-maker OpenAI race to sell shares on the stock market, with an IPO that could value Anthropic at nearly ₹8 lakh crore. Anthropic co-founder Jack Clark and research institute head Marina Favaro stated that the pause would enable 'societal structures and alignment research' to keep up with AI advances, ensuring the technology matches human values and intentions. The proposed coordination would allow advanced AI labs to verify that global rivals have actually stopped or slowed their work, preventing bad actors from using the auspices of a coordinated slowdown to jump ahead in secret.
The ownership plan lands at a sensitive moment for both companies, with Anthropic submitted a confidential S-1 on June 1 after a $65 billion Series H valued it at $965 billion, while OpenAI was last valued at $852 billion in March and is preparing its own listing. According to CNBC, the administration has precedent for taking stakes, with a similar stake at OpenAI's valuation shifting roughly $85 billion away from existing shareholders and IPO buyers. Political pressure spans both parties, with Senator Bernie Sanders having proposed a one-time 50% tax, paid in shares, on OpenAI, Anthropic, and xAI. Investors weighing the $3 trillion IPO wave must therefore price governance risks that Anthropic does not currently carry, as the company's independence from Washington could now read as a cleaner ownership story for public investors.
According to a comprehensive analysis by Anthropic, artificial intelligence is fundamentally transforming cybersecurity threats in 2026. The company examined 832 accounts that were banned for malicious cyber activity between March 2025 and March 2026, revealing significant shifts in how threat actors are leveraging AI technology. As reported by Anthropic, threat actors are no longer using AI only for basic cybercrime activities such as writing malware or phishing emails, but are increasingly employing AI for more sophisticated operations after gaining system access. The analysis covered 13,873 distinct malicious actions and observed 482 unique MITRE ATT&CK techniques across the dataset.
The analysis revealed significant gaps in current cybersecurity methodologies. According to Anthropic, existing frameworks like the widely used MITRE ATT&CK framework do not fully account for AI-enabled attack techniques. The company cited a state-backed cyber espionage operation from 2025 where an AI agent executed commands, exploited vulnerabilities, stole credentials, and made tactical decisions with minimal human intervention. Anthropic noted that there is no ATT&CK ID for agentic orchestration behaviors, despite these being precisely the techniques expected to become more prevalent as AI agents become more capable. The company is in talks with MITRE about extending ATT&CK to cover AI-enabled behavior, and some of this analysis already fed into Verizon's 2026 Data Breach Investigations Report. Researchers have also warned that 'it's not just the biggest, most powerful language models that pose the security concerns', as even open-source AI tools can be easily accessed and modified to create sophisticated cyber threats.
