#245: OpenAI DevDay and Dots, OpenAI’s Safety Crisis, Trump Renames Artificial Intelligence & Anthropic Targets Pre-Thanksgiving IPO Ad:
The Artificial Intelligence Show
20 HOURS AGO
#245: OpenAI DevDay and Dots, OpenAI’s Safety Crisis, Trump Renames Artificial Intelligence & Anthropic Targets Pre-Thanksgiving IPO Ad:
#245: OpenAI DevDay and Dots, OpenAI’s Safety Crisis, Trump Renames Artificial Intelligence & Anthropic Targets Pre-Thanksgiving IPO Ad:
The Artificial Intelligence Show
20 HOURS AGO
Paul Roetzer and Mike Kaput unpack OpenAI’s DevDay and safety troubles, the politics of “superintelligence,” Anthropic’s IPO ambitions, new liability pressures, emerging AI capabilities, practical use cases, and the week’s product news.
This episode examines a rapidly changing AI industry where product releases, safety concerns, political influence, and commercial pressure increasingly collide. The hosts assess OpenAI’s Dots agents and DevDay announcements, question whether frontier labs can maintain credible safety practices, and scrutinize the White House’s voluntary Super Intelligence Accord and renaming campaign. They also explore Anthropic’s potential IPO and political exposure, Google’s Gemini 4 preview, FTC scrutiny and agent liability, debates over AI consciousness, and Tavis’s realistic real-time video model. The episode closes with examples of AI-assisted creative work and updates on products, privacy, funding, and upcoming events.
05:10
05:10
The episode starts with OpenAI
10:07
10:07
DevDay’s reaction was pretty meh
14:48
14:48
Personal agents may crater after the novelty wears off
25:04
25:04
Iterative deployment guarantees periodic failures
29:14
29:14
The Accord Is Voluntary
35:33
35:33
AGI Is Already Here, Unevenly Distributed
40:37
40:37
Rebranding is not superintelligence
48:04
48:04
Passing the loyalty test failed the backbone test
51:11
51:11
A name change without a definition
55:05
55:05
Anthropic could be valued above $2 trillion
59:23
59:23
Argon expands the context window to one million tokens
1:05:26
1:05:26
The investigation may be a warning shot
1:11:45
1:11:45
AI can appear conscious without being conscious
1:18:11
1:18:11
AI video is becoming nearly indistinguishable from a human conversation
1:23:58
1:23:58
AI using AI feels a little sci-fi
1:30:13
1:30:13
Apple tightens access controls because of autonomous agents
