Zoho founder Sridhar Vembu
Because many of those predictions have proved more accurate than he expected, maynard said he had been sceptical of earlier AI predictions that sounded like science fiction but now takes the risks more seriously.
Sridhar Vembu said he has not bought into AI doom scenarios, but believes people should remain open to changing their views as AI models become more capable. Sridhar Vembu said people need both personal conviction and scientific scepticism when thinking about AI and its future. “I have not bought into the AI doom scenarios,” Vembu wrote on X, pointing to Maynard’s changing position as an example of what he described as a scientific attitude. Maynard said he had followed AI for some time but had been sceptical of predictions that appeared to come from science fiction. Vembu said personal conviction helps people continue working toward a goal, while scientific scepticism can prevent people from becoming disconnected from evidence. “Our own hunches and convictions allow us to pursue and stick to a course of action and scientific scepticism keeps us grounded in reality,” he wrote.
Zoho founder and chief scientist Sridhar Vembu has shared his view on the growing debate over whether AI could pose serious risks to humanity. Vembu was responding to a post by American physicist Steve Hsu, who shared comments from Fields Medalist James Maynard on X. “Fields Medalist James Maynard teaches us the scientific attitude about this: he has been highly sceptical of claims of AI doom but as the models demonstrate higher and higher capability levels, he is taking them more seriously now. Steve Hsu’s post quoted Maynard discussing how his view of AI risks has changed. “I’ve been interested and I’ve been following AI, but I’ve heard all these predictions that just sound like sci-fi and I’ve been consistently dismissive and skeptical of all of them. “But I’ve been consistently wrong. The people who sound like they’re talking from a sci-fi novel are the people who’ve been making much much more accurate predictions. Maynard concluded that AI risks should not simply be dismissed as hype. He added: “Without a strong internal conviction, our actions lack the will power to succeed. Without scientific scepticism, we risk being delusional. Get the latest technology news and updates. Download the TOI App.
He then acknowledged that his earlier view had been wrong.