Bill Gates says it’s ‘not enough to have a kill switch’ for AI: Full interview

Summary
In an NBC News Meet the Press interview, host Kristen Welker interviews Microsoft co-founder and philanthropist Bill Gates about the trajectory, governance, and societal risks of artificial intelligence. Gates discusses the Gates Foundation’s global initiative to expand AI language equity, addresses existential and near-term threats including bioweapons and cyberattacks, rejects simple solutions like a “kill switch,” calls for binding government regulation, and comments on his past meetings with Jeffrey Epstein as well as AI’s future impact on education and healthcare.

What is shown

• [00:00] Kristen Welker introduces Bill Gates on Meet the Press in a studio setting overlooking city high-rises.
• [00:39] Gates discusses language disparity in existing AI models and outlines the Gates Foundation’s collaborative initiative with roughly 60 organizations.
• [02:55] Gates recounts OpenAI demonstrating GPT capabilities on an Advanced Placement Biology exam roughly six months prior to public release.
• [06:09] Gates addresses questions regarding human control over AI systems and current computational limits.
• [07:59] Gates discusses severe near-term risks involving malicious actors employing AI for cyberattacks and bioweapons.
• [10:47] Gates analyzes regulatory proposals, critiquing simplistic “kill switch” concepts and urging mandatory, verifiable monitoring safeguards.
• [17:15] Gates discusses legislative urgency in Washington, international coordination, and diplomatic engagement with China.
• [21:44] Gates responds to political characterizations of existential AI risk as a “hoax” and discusses potential bilateral US–China guardrails.
• [24:52] Welker questions Gates regarding his past meetings with Jeffrey Epstein and his testimony before Congress.
• [27:19] Gates discusses the role of AI in education, cautioning against sycophantic conversational companion bots for children while praising adaptive tutoring tools.
• [28:13] Gates outlines how he uses AI tools to synthesize research on malaria and malnutrition for the Gates Foundation.

Claims & numbers

• Gates states that for African languages such as Yoruba, AI error rates are roughly 10 times higher than for English.
• Gates notes the Gates Foundation’s initiative partners with approximately 60 organizations to bring AI capabilities to every language spoken by 1 million or more people within a 5-year timeline.
• Gates claims that AI in the hands of bad actors has the potential to trigger catastrophic events causing up to “a billion deaths.”
• Gates emphasizes that industry self-regulation is insufficient and that government mandates and auditing safeguards are urgently necessary.
• Gates notes that the Gates Foundation plans to spend over $200 billion over the next 19 years, using AI to improve outcomes in global health areas like malaria and malnutrition.

Notable quotes

• [08:00] “AI is certainly powerful enough to drive events that, you know, cause a billion deaths.”
• [13:15] “It’s not enough to have a kill switch... You need insight and records of what’s being done.”
• [15:15] “No one thinks self-regulation is enough.”

Assessment
This is a genuine, professionally recorded broadcast news interview conducted on set by NBC News. No technical demonstrations or synthetic media are presented; the video consists entirely of a standard two-person journalistic dialogue.

Described by gemini-3.8-flash on 2026-10-08 from the video’s audio and frames.