In a recent a16z podcast, OpenAI CEO Sam Altman expressed concerns about the potential dangers of artificial intelligence (AI), particularly with the launch of Sora 2, a new video application. Altman warned of “really strange or scary moments,” highlighting the rapid creation of deepfake videos, including some featuring Dr. Martin Luther King Jr., which prompted an apology from OpenAI. He emphasized the urgent need for society and AI to “co-evolve,” cautioning against letting computers make decisions independently. While Altman acknowledged that catastrophic AI events haven’t yet occurred, he believes they could happen, potentially resulting in significant societal shifts from widespread AI interactions. Despite these risks, he opposes heavy governmental regulation, suggesting that careful safety testing is preferable for advanced AI models. Altman concludes that societal adaptation will foster necessary guardrails for AI technology, reinforcing the need for responsible innovation as AI continues to evolve.
Source link
