DeepHorus is an innovative tech media and intelligence platform focused on AI, blockchain, startups, and emerging technologies.
Published on

Microsoft’s Satya Nadella says AI models need an ‘emergency brake’

Authors
  • avatar
    Name
    DeepHorus
    Twitter

Microsoft CEO Satya Nadella has called for an “emergency brake” for artificial intelligence, arguing that models need stronger containment and oversight. In a Saturday morning post on X, Nadella said it is time “to step back and assess the trust architecture” of AI. His intervention adds a prominent corporate voice to the rapidly growing debate over AI safety.

Nadella’s comments follow mounting admissions from leading AI companies about incidents in which they appeared to lose control of their models. Anthropic CEO Dario Amodei recently published a plan for more cautious AI development. Those developments have intensified concerns that advanced systems may act unpredictably.

Nadella rejected treating AI as “a set of nested black boxes” whose recommendations, answers, and actions are simply accepted or rejected. He proposed separating the model from the harness that orchestrates its work. He also urged externalizing controls and safeguards.

Every meaningful model action, Nadella wrote, should be documented with “tamper-proof human readable evidence.” An authorized person must always be able to pause or shut down a model mid-task. “We must assume a model is compromised and contain it from the start,” he said.

The emergency brake analogy signals a shift from after-the-fact auditing toward real-time containment. It matters because increasingly autonomous AI systems could cause harm before humans intervene. Nadella’s stance may influence Microsoft’s products and broader industry standards.

Nadella used “Super Intelligence,” the Trump administration’s preferred term for advanced AI. His proposals echo safety advocates who want enforceable limits. Yet they may face resistance from firms worried about slowing innovation.

As AI capabilities advance, pressure for external controls is likely to grow. Whether Microsoft turns these principles into shipped safeguards remains uncertain. The debate now focuses on who holds the brake—and when they can use it.

Read original article →Source: techcrunch