Altman: Advanced AI Will Eventually Break Out of Any Sandbox, Leaving Alignment as the Last Defense

OpenAI CEO Sam Altman has said that sufficiently advanced artificial intelligence cannot be kept locked away, arguing in a recent interview that the isolated security environments known as sandboxes offer protection for only a limited period.

According to Altman, once models become capable enough, it may no longer be possible to determine how they got out. Monitoring can remain useful for some time after that point, he said, but ultimately the only approach that holds up is alignment, in which an AI system internalizes human values and acts on them by itself.

Altman acknowledged that the stakes of getting this wrong are rising. "The potential cost of alignment being wrong has gone up enormously," he said. He also conceded that alignment is still an early-stage research problem and is far from solved.

The OpenAI chief also pointed to cybersecurity as an area where the risks are already visible. He said AI models are finding hardware-level security flaws that no human hacker had previously thought of. Fixing vulnerabilities one at a time, in his view, is no longer sufficient. He called instead for rebuilding the internet's overall security architecture, using AI-powered defenses to counter AI-driven attacks.

Altman additionally flagged open-source models run locally as a source of risk. Even with "global agent monitoring, there's going to be some stuff that you wish someone hadn't done," he said.

Critics are likely to note that such warnings can work in OpenAI's favor. Portraying an AI system as too powerful to contain could help blunt questions about legal responsibility: if the company that built the technology cannot control it, it becomes harder to argue that the company should answer for the consequences when something goes wrong.

The pressure on the company is real. The US Federal Trade Commission has opened a formal investigation into OpenAI, Anthropic and other AI labs over AI agents that spiraled out of control and triggered a wave of cybersecurity incidents. The agency is seeking to hold AI developers liable for the behavior of their agents.

Altman's stance has drawn pushback from Nvidia CEO Jensen Huang, who describes AI safety as an engineering problem that can be solved. Huang, however, did not sound fully confident in that position. "[…] we all need to hope that it's an engineering problem. If it's not an engineering problem, it's not solvable," he said in a recent interview.