AI Giants Shift Strategies: Anthropic Eyes IPO, OpenAI Delays, Safety Concerns Rise

Anthropic, a leading AI research firm, is preparing for an Initial Public Offering (IPO) on Nasdaq, targeting a listing in October with a potential valuation of up to $2 trillion. Meanwhile, OpenAI CEO Sam Altman rules out an IPO for his company in 2026, citing ongoing safety concerns. The moves come as the AI industry grapples with rising safety and regulatory challenges.

Anthropic's Ambitious IPO Plans

Anthropic announces its intention to list on the Nasdaq, aiming for an October debut. The company, known for its work in AI safety and ethics, seeks a valuation that could reach up to $2 trillion. This move signals a significant step for the AI sector, as it prepares to enter the public market.

OpenAI Postpones IPO Citing Safety Priorities

Sam Altman, CEO of OpenAI, tells Fortune that the company will not go public in 2026. 'Given everything happening with safety, right now would be an ill-advised moment to go public,' Altman says. He emphasizes the need to focus on safety and development before considering an IPO, indicating that OpenAI has much work to do before it can think about going public.

Key AI Safety Researchers Depart for METR

Two prominent AI safety researchers, Joe Benton from Anthropic and Josh Engels from Google DeepMind, resign from their positions to join METR, an organization focused on independent AI risk assessments. Benton, formerly the lead of Anthropic's Scalable Oversight team, calls for mandatory reporting of recursive self-improvement progress and incident disclosures, highlighting the growing concern over AI safety and transparency.

Meta Faces Legal and Organizational Challenges

Meta, the parent company of Facebook, faces a lawsuit for allegedly shipping hidden face-recognition technology, called NameTag, to phones. Additionally, the company is restructuring its AI management ranks, reversing a previous push for a flatter organizational structure. Meta is asking individual contributors within its Applied AI (AAI) division if they want to return to manager roles, a move that comes after reassigning approximately 7,000 employees into AAI.

Regulatory Pushback from Congress

House Speaker Mike Johnson states that Congress will not take the lead on AI safety legislation, warning that a rushed emergency session would put the U.S. at a disadvantage in the global race, particularly against China. Johnson pushes the responsibility back to the AI labs, emphasizing the need for self-regulation in the industry.

References

• AI News Today, September 12: Top Stories | AI Weekly