Sam Altman Shows Me GPT 5... And What's Next

The interview offers a rare opportunity to hear directly from Sam Altman about GPT-5 and OpenAI’s future plans. However, the value of the information is undermined by Altman’s tendency to avoid direct answers, particularly on ethical and societal concerns. His responses often pivot to optimistic generalities or product promotion, leaving critical questions unaddressed. For instance, when asked about how we will know what’s real in 2030, he sidesteps the issue, focusing instead on the benefits of AI. This pattern is consistent throughout the interview, leading to a sense that the conversation is more of a PR exercise than a substantive discussion. The scientific rigor is limited; while Altman mentions technical milestones like IMO gold medal performance, he provides little detail on the underlying methods or limitations. The single source cited in the description is an IMDB page, which is irrelevant to the content, indicating a lack of supporting references. The interview does offer some insights into OpenAI’s thinking, such as the emphasis on coding as a key capability and the prediction of AI-driven scientific discoveries within two years. However, these claims are not backed by evidence or independent analysis. The title accurately reflects the content, but the substance is lacking. Overall, the interview is informative from a corporate perspective but fails to provide a balanced or rigorous examination of the implications of AI. The public comments reflect widespread skepticism about Altman’s evasiveness, with many viewers noting that he rarely answers questions directly. This suggests that the interview may not satisfy those seeking critical analysis. In conclusion, while the interview provides a platform for Altman to share his vision, it falls short in delivering the depth and accountability expected from a scientific discussion.

The title accurately reflects the content: an interview with Sam Altman about GPT-5 and future plans.

6/10

The interview features Sam Altman, CEO of OpenAI, discussing GPT-5 and future AI developments. While the content is informative and provides insights into OpenAI's perspective, it lacks independent verification and often avoids direct answers to critical questions. The single source cited is an IMDB page, which is not relevant to the technical claims. The discussion is more promotional than rigorously scientific.

• 00:00What future are we headed for?
• 02:06What can GPT-5 do that GPT-4 can’t?
• 06:57What does AI do to how we think?
• 10:52When will AI make a significant scientific discovery?
• 13:09What is superintelligence?
• 16:17How does one AI determine “truth”?
• 18:35It’s 2030. How do we know what’s real?
• 21:20It’s 2035. What new jobs exist?
• 24:02How do you build superintelligence?
• 26:00What are the infrastructure challenges for AI?
• 28:18What data does AI use?
• 29:50What changed between GPT1 v 2 v 3…?
• 32:55What went right and wrong building GPT-5?
• 35:40“A kid born today will never be smarter than AI”
• 37:57It’s 2040. What does AI do for our health?
• 40:00Can AI help cure cancer?
• 41:10Who gets hurt?
• 43:00“The social contract may have to change”
• 45:22What is our shared responsibility here?
• 49:21“We haven’t put a sex bot avatar into ChatGPT yet”
• 51:40What mistakes has Sam learned from?
• 53:10“What have we done”?
• 57:40How will I actually use GPT-5?
• 59:40Why do people building AI say it’ll destroy us?
• 1:03:16Why do this?

• Sam Altman - IMDb — The only link provided in the video description, likely for Sam Altman's IMDb page, though not directly relevant to the technical content.

• OpenAI official website — Provides official information about GPT-5 and OpenAI's mission, aligning with the video's claims.

• Critiques of AI hype — Many public comments and external analyses argue that Altman's claims are exaggerated and that the risks of AI are downplayed, contrasting with the optimistic tone of the interview.