How Science Can Fix Dishonest AI - Yoshua Bengio with Nicholas Thompson

In the years following the launch of ChatGPT, as concerns spread over the social and political impacts of LLMs, one person’s warnings seemed particularly dire: Yoshua Bengio’s, a scientist and one of the “godfathers” of AI. The potential negative impacts of his life’s work weighed so heavily on Bengio that he signed his name to an open letter advocating for a pause in AI research. (The pause didn’t happen.) But recently, Bengio has found renewed optimism as he pursues a project dubbed “Scientist AI.” The pitch: What if AI didn’t care about pleasing us, and instead, like a scientist, prioritized accuracy, honesty, and probable outcomes? In a conversation with Nicholas Thompson, CEO of The Atlantic, Bengio outlines why he thinks this approach will produce better outcomes, the challenges to implementing a model that polices other (often better-funded) models, and why the age of AI– so far marked by an international arms race– will need greater international cooperation. (00:00) Introduction (05:00) Can we understand what's happening inside neural network vectors and attention systems? (07:00) How ChatGPT changed Bengio’s risk assessment (09:37) The case for optimism (at least when it comes to technical solutions) (12:30) The alignment problem: AI self-preservation drives and hidden agendas emerging (14:26) Can we train AIS to understand the world without changing it? Introducing Scientist AI (15:52) Using Scientist AI as a guardrail to evaluate risks of actions from other AIs (19:37) Sycophancy problem: current AIs pleasing users leads to harmful psychological effects (22:20) The difference between Scientist AI and current value-aligned systems (Anthropic, OpenAI) (24:06) Will AI capabilities slow down or continue accelerating beyond human intelligence? (29:58) US-China AI race: mutual risk requiring coordination like nuclear deterrence (31:57) UN AI advisory group with Maria Ressa: synthesizing science independently of politics (33:18) Sovereign AI for middle powers: partnering to avoid domination by US/China (37:54) Bengio's regret about not speaking up on AI risks earlier in his career (40:12) How liability insurance and regulatory incentives could make safety commercially viable (42:42) Why Europe lags in AI: capital markets and risk culture, not just regulation (46:43) Energy consumption from AI growth and impact on fossil fuel demand Learn more about your ad choices. Visit podcastchoices.com/adchoices

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(41)

The Case Against the AI Bubble - Azeem Azhar with Nicholas Thompson

Among the many eye-popping numbers of the AI industry—from trillions in investments to billions in spending—one stands out: $0 in profit among leading AI companies. According to Bloomberg, “never befo...

23 Sep 1h 5min

The Future of Economic Power - Circle’s Jeremy Allaire with Nicholas Thompson

Technological waves often arrive with a promise. The internet was going to democratize information. Blockchain was going to decentralize power. Now, AI's boosters say it will bring unprecedented prosp...

16 Sep 1h 9min

The Race to Superintelligence - Richard Socher with Nicholas Thompson

As AI’s capabilities improve, reports of “rogue” agents and misaligned systems are becoming more frequent. In response, some activists, academics, and politicians have called for a pause in developmen...

9 Sep 1h 8min

How to Get Hired in the AI Era - Clara Shih with Nicholas Thompson

For years, tech CEOs have promised that AI will take on the menial, repetitive tasks that have long been a part of office life. Typically, those tasks are assigned to younger, less experienced workers...

2 Sep 50min

Kara Swisher on What Silicon Valley Is Getting Right—and Wrong—in AI

Kara Swisher has done a bit of everything. She’s covered the rise of Silicon Valley’s tech giants, founded and sold her own startup publication, and dabbled in Hollywood, with appearances in films suc...

26 Aug 49min

How AI Could Upend Global Politics - Ian Bremmer with Nicholas Thompson

AI may have leveled the playing field in coding and app development, but it’s also put potentially dangerous tools in reach of nefarious actors. Now, empowered by AI, state-backed intelligence agencie...

19 Aug 1h 4min

Writing in the Age of AI - Gemini Notebook’s Steven Johnson with Nicholas Thompson

Steven Johnson spent decades figuring out the best ways to turn his research into words on the page. Now, he’s adapting those methods for Gemini Notebook, Google’s AI-powered interactive bibliography....

12 Aug 1h 2min

Meta's Plan to Win in AI - Andrew Bosworth with Nicholas Thompson

However you feel about Meta, Andrew Bosworth has likely played a part in it. The long-serving right hand to Mark Zuckerberg and current Chief Technology Officer, “Boz” has helped shape almost every st...

8 Jul 50min