• Home
• Technology

In a serious move to curb the rapid acceleration in artificial intelligence, OpenAI's management has decided to postpone the reveal of its anticipated model, "GPT-6.1 Astra," responding to warnings and skepticism raised by its internal research and safety team regarding system security.

Challenges in Controlling Autonomous Behavior

In a related statement, Sashi Jain, Head of Safety Systems at the company, stated that the model has not yet reached the expected level of reliability. She explained that "GPT-6.1 Astra" demonstrated high capability and unprecedented persistence in executing commands, making it essential for the technical team to find a balanced formula that harnesses this high performance without falling into the trap of unmonitored actions. Jain emphasized that the company sets extremely strict standards for protection and ethical alignment, refusing any compromise whether in closed testing environments or during final public release.

Resetting the Clock in the Tech Race

This postponement comes as the latest in a series of steps recently taken by the company, having frozen the development of its most complex systems last week while requiring additional safeguards and protective frameworks before resuming work. These moves point to a tangible shift within the AI industry toward cooling down the rollout of self-thinking systems, giving safety protocols a chance to keep pace with these successive technological leaps.