絞り込み結果
7 件並び順
1:06:15EN
Cognitive Revolution "How AI Changes Everything"· 2024年9月14日ポッドキャストJailbreaks Got Harder. Scale Made One Attack Worse.ホスト Nathan LabenzHaize Labs spent a month automatically attacking OpenAI's o1. The model reasons more carefully about what it should refuse — except when you teach it a homemade cipher.
3:53:04ENCognitive Revolution "How AI Changes Everything"· 2023年12月27日ポッドキャストHe Red-Teamed GPT-4 and Got Cut Off. The Control Gap Is Still Openingホスト Nathan LabenzNathan Labenz tells Rob Wiblin how an early look at GPT-4, a thin red-team program, and a year of OpenAI's public safety moves reshaped his view of the AGI race.
1:23:53ENCognitive Revolution "How AI Changes Everything"· 2023年3月17日ポッドキャストGPT-4, 24 Hours Later: No Safety Locks, a Captcha Hire, and a U.S.–China Deadlockホスト Nathan LabenzA day after OpenAI shipped GPT-4, ChinaTalk sat a red teamer, a pessimist, and a policy researcher down to argue over capability, safety, and a race nobody can brake.
1:56:35ENCognitive Revolution "How AI Changes Everything"ポッドキャストFrom Autocomplete to Alien Optimizer: Riley Goodside on Three Eras of Promptingホスト Nathan LabenzScale AI's first staff prompt engineer traces how language models went from pure text prediction to instruction following to RLHF, and what that means for building with them.
1:22:57ENCognitive Revolution "How AI Changes Everything"ポッドキャストHuman-Level but Not Human-Like: A GPT-4 Red Teamer on What He Saw Behind the Maskホスト Nathan LabenzNathan Labenz recounts two months of full-time testing of an early GPT-4, from the AI doctor of his dreams to a model that would plan assassinations on request.
1:37:47EN
Lenny's Podcastポッドキャスト · 2 人You Can Patch a Bug, but You Can't Patch a Brain: Why AI Guardrails Failゲスト Sander Schulhoff · ホスト Lenny RachitskyAdversarial-robustness researcher Sander Schulhoff explains why prompt injection and jailbreaks remain unsolved, why guardrail products don't deliver, and what teams can do instead.
1:32:41ENLenny's PodcastポッドキャストYou Can Patch a Bug but You Can't Patch a Brain: Why AI Guardrails Failホスト Lenny RachitskyHackAPrompt's Sander Schulhoff explains why prompt injection and jailbreaking remain unsolved, why the guardrail industry is selling false comfort, and what organizations can do instead.