Q&A with Google Chief Internet Evangelist Vint Cerf, who was awarded IEEE's Medal of Honor, on how Google has changed since 2005, the hazards of LLMs, and more
Context & Ripple Effects
TechCrunch's interview with Vint Cerf lands at a moment when Google's own leadership is defining its AI posture: Sundar Pichai has been making the case for Google's place in the AI race in his own Q&A on Search's future and public skepticism of AI. Cerf, Google's chief internet evangelist since 2005 and now an IEEE Medal of Honor recipient, uses the platform to flag the hazards of LLMs rather than to sell them.
The piece reads differently in hindsight given the related coverage: Cerf went on to plan his retirement from Google in 2026, making this one of his substantive late-career statements on how the company changed since he joined — and on the technology now reshaping it.
First-order effects
- Cerf's caution about LLM hazards, coming from a Medal of Honor-level figure inside Google, gives external critics of large language models an authoritative internal citation point.
Second-order effects
- The interview sharpens the contrast between Cerf's hazard framing and Google's aggressive AI push under Pichai, forcing the company's public narrative to accommodate a respected internal skeptic.
Third-order effects
- If pioneers-turned-elders keep voicing structural caution as their companies deploy LLMs, the industry's legitimacy argument shifts from 'can it work' to 'who is accountable for its failure modes' — a framing regulators and standards bodies like IEEE are positioned to absorb.
The trend: As AI becomes Google's defining product bet, its own veteran technologists are emerging as prominent voices on LLM risk, signaling a widening gap between deployment pace and internal confidence in safeguards.