Barry Diller Defends Altman but Warns About the Danger of AGI

Barry Diller defended Sam Altman, but with a caveat that sounds like a warning: trust in a person is irrelevant once AGI is on the horizon.

Defense and Alarm at Once

The media mogul and founder of IAC did not criticize the OpenAI CEO. Diller acknowledged Altman's competence. But he then added a blunt caveat: as AGI approaches, a leader's competence is not the answer. Trust in a person becomes irrelevant in the face of a technology that is hard or impossible to fully control.

Why Trust Is Irrelevant

AGI is a hypothetical intelligent system that would surpass human intelligence in every domain. If AGI does arrive, the motives, competence and honesty of any one individual will prove an insufficient safeguard. A system that outstrips the human mind can slip out of control regardless of its creators' intentions.

• Even a conscientious leader cannot guarantee the safety of AGI
• The problem lies in the nature of AGI itself, not in the quality of management
• Technical constraints built into the system itself are required

"Trust is irrelevant when it comes to AGI" is the gist of Diller's position, although he does not deny Altman's professionalism.

Guardrails as the Only Way

Diller stresses the need for "guardrails": control and restriction mechanisms that must be built into AGI at the level of architecture and algorithms. This is not a question of management or of company culture. It is a question of technical safety. Such mechanisms must prevent the system from drifting away from its intended goals, no matter who controls it. Guardrails work automatically, at the level of code and algorithms, not at the level of human decision.

What It Means

Diller's statement reflects a growing consensus in the AI community: preparing for AGI means not picking good people for leadership positions, but building technology that is safe by default. Protection against AGI is not a matter of trusting Sam Altman or any other leader. It is a matter of engineering.