OpenAI, Anthropic, and Google Negotiate Unprecedented AI Safety Standards Body
In what may be the most consequential governance development in the AI industry this year, OpenAI, Anthropic, and Google have confirmed they are in active discussions to create a shared AI safety standards body. The talks, which have been ongoing since at least July, center on a proposal by Google DeepMind CEO Demis Hassabis for a self-regulatory organization modeled after the Financial Industry Regulatory Authority (FINRA).
The proposed body would test the most powerful AI systems before they are released to the public — a significant step beyond the voluntary commitments the companies have made in the past. At a company-wide town hall, OpenAI CEO Sam Altman acknowledged that the major AI labs would likely have to build the standards body themselves, without relying on the U.S. government to take the lead.
No agreement has been reached on what the standards would actually require, and the companies remain fierce competitors. But the fact that all three are at the table signals growing recognition that frontier AI capabilities — especially after the recent wave of cyber-capable models — demand coordinated oversight that goes beyond any single company's internal safety team.
Sources: CNBC, TechCrunch, Seoul Economic Daily
OpenAI Seeks $1.5 Trillion Valuation, Pushes IPO to 2027
OpenAI is weighing a new funding round at a valuation of $1.2 to $1.5 trillion, according to reports from Bloomberg and the Financial Times. If confirmed, this would make OpenAI more valuable than nearly every public company on Earth — a staggering trajectory from its $157 billion valuation just over a year ago.
The fundraising push comes as the company reports more than $40 billion in annualized revenue, roughly double its sales at the end of 2025. Growth has been driven by strong enterprise adoption of its Codex coding tool and the September launch of GPT-6 Astra, the first model to trigger OpenAI's critical-cyber safeguard threshold.
Meanwhile, CEO Sam Altman has pushed the company's IPO timeline to 2027, citing growing concerns about existential risks from AI. The delay is notable: even the most commercially aggressive AI lab is invoking safety as a reason to manage its public market timing.
Cohere and Aleph Alpha Finalize $20 Billion Sovereign AI Merger
On September 16, Cohere and Aleph Alpha signed a definitive business combination agreement, formalizing a deal first disclosed in April. The unified company will be dual-headquartered in Berlin and Toronto, valued at roughly $20 billion, and positioned as the first transatlantic sovereign AI provider for governments and regulated industries.
Germany's Schwarz Group — owner of Lidl and Kaufland — is committing $600 million to lead the combined firm's Series E round. Both the Canadian and German governments have backed the deal, each contributing up to $150 million in grants.
The merger validates the "sovereign AI" thesis: that nation-states and large enterprises will pay a premium for AI systems that are not routed through U.S. hyperscalers. By combining Cohere's enterprise NLP platform with Aleph Alpha's deep ties to European government and defense clients, the new entity offers a credible alternative to the Big Tech AI stack.
Sources: Unite.AI, TechCrunch
Chinese AI Chip Maker Biren Eyes $1 Billion Share Sale After 2,000% Revenue Surge
Shanghai Biren Technology is considering raising approximately $1 billion in a new share placement — its third capital raise since its January 2026 Hong Kong IPO. The company posted first-half 2026 revenue of $183.9 million, up nearly 2,000% year-over-year, driven by surging demand for domestic AI chips as U.S. export controls push Chinese firms away from Nvidia and AMD.
Biren has become a bellwether for China's AI chip self-sufficiency push. The explosive growth demonstrates that U.S. export controls are actively creating a domestic Chinese AI chip market rather than simply constraining capabilities. The third billion-dollar raise in under nine months reflects strong investor conviction that Chinese AI infrastructure demand will remain structurally high, regardless of Western chip restrictions.
California Signs Strictest Child Safety Laws for AI Chatbots
Governor Gavin Newsom signed a suite of laws described as the strongest child safety chatbot and social media regulations in the nation. The legislation requires AI chatbot companies to conduct risk assessments before new rollouts, include crisis protocols for suicidal ideation, expand parental controls, and submit annual independent child safety audits.
Companies face penalties of up to $1 million per child if found guilty of causing harm. The laws also prohibit "addictive features" for users under 16, targeting the rapidly growing AI companion chatbot category.
Given California's market size, these regulations effectively set a de facto national standard. Every major chatbot provider will need to build dedicated child safety infrastructure — a requirement that could reshape how consumer-facing AI products are designed across the United States.
Sources: Office of Governor Newsom, EdSource