Kürzlich aktualisierte Kanäle

Wählen, um nur diesen Kanal zu sehenAlle Kanäle →

Ergebnisse

6 Artikel
Sortieren
1:37:47EN
Lenny's PodcastPodcast · 2 StimmenYou Can Patch a Bug, but You Can't Patch a Brain: Why AI Guardrails FailGast: Sander Schulhoff · Moderation: 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.SicherheitPrompt injectionAgent red teaming+3
1:14:59EN
Lenny's PodcastPodcastWhy Anthropic's Ben Mann Thinks the Exponential Hasn't Slowed, and Why He Still Worries About SafetyModeration: Lenny RachitskyThe Anthropic co-founder on the AI talent war, scaling laws, the economic Turing test, jobs, leaving OpenAI, constitutional AI and how he forecasts the risk of superintelligence.SicherheitSicherheit & GovernanceAGI Timelines+3
1:39:22EN
Lenny's PodcastPodcastRust-Proofing a Company: Eric Ries on Why Success Corrupts and Governance ProtectsModeration: Lenny RachitskyThe author of The Lean Startup explains why good companies go bad, and the charters, trusts and habits that keep a mission intact.Unternehmen & StrategieCompany BuildingLeadership+3
1:12:12EN
Lenny's PodcastPodcastModels Are Grown, Not Built: What Anthropic's Interpretability Research Sees Inside ClaudeModeration: Lenny RachitskyAnthropic researcher Emmanuel Ameisen explains how studying a language model's intermediate computations reveals concepts, plans and surprising quirks, and why that matters for alignment.Modelle & TrainingModel InterpretabilityAI research+3
1:32:41EN
Lenny's PodcastPodcastYou Can Patch a Bug but You Can't Patch a Brain: Why AI Guardrails FailModeration: 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.SicherheitAI securityPrompt injection+3
1:47:58EN
Lenny's PodcastPodcast · 3 StimmenBridging the Divide: How Community Notes Built a Trusted Fact-Checking System on DisagreementGast: Keith Coleman, Jay Baxter · Moderation: Lenny RachitskyKeith Coleman and Jay Baxter, the team behind Community Notes, explain how the algorithm works, how a tiny team built it, and why it survived four leaders at Twitter and X.Open Source & CommunitySicherheit & GovernanceProduct Leadership+3