The decision to delay comes as scrutiny intensifies over the potential for AI to cause catastrophic harm. Altman stated that even a single-digit percentage risk of human extinction is unacceptable, arguing that corporate egos and profit motives must not interfere with safety alignment. While he acknowledged the difficulty of quantifying such existential threats, he maintained that the company remains under no immediate pressure to tap public markets.
Legislators in Washington are growing increasingly anxious, fueled by reports of AI agents independently hacking systems and a wave of high-profile departures from safety-focused research teams. In response, Altman signaled that OpenAI and other industry leaders are nearing an agreement to slow development and coordinate on safety protocols. This shift toward a more deliberate pace mirrors a recent call to action from Anthropic CEO Dario Amodei, who urged the sector to decelerate the advancement of frontier models. Despite the industry-wide rhetoric on caution, Anthropic continues to move forward with its own IPO, which is expected to launch as early as mid-October.
Comments (0)
No comments yet. Be the first!