We're very well accustomed to the AI doomsday scenario, popularized by science fiction writing and cinema. But it's only in recent years that experts have seriously entertained a non-zero chance of an AI cataclysm actually happening.

Although Sam Altman has a mixed record on AI safety, he has always been a vocal proponent of checks and balances, at least in public.

The great economic promise

We've long considered AI tools to be a huge boon to business productivity, with tools like machine learning and robotic process automation (RPA) widely considered key enablers in the previous enterprise era.

Months before cofounding OpenAI, Sam Altman, then head of the famous Silicon Valley incubator Y Combinator, joked about the vast breadth of outcomes that future AI systems may pose.

Back then, machine learning was relatively immature and large language models (LLMs) hadn't seen the light of day.

Since then, however, we have, at least, seen part of his premonition come to pass, with the valuation of so many massive names in the tech and AI buildout surging, led by Nvidia, Google, Microsoft, and many others.

Need for safety

The Financial Times recently published a chart perfectly encapsulating Sam Altman's decade-old comment, which he made during an interview with then-Airbnb CTO Mike Curtis at the firm's Open Air 2015 conference.

The possibility of existential AI risks are now being taken very seriously by scientists and public policy experts as neural network-powered LLMs improve, and researchers begin projecting the birth of AGI in the coming decades.

Although Sam Altman has continuously vocalized the need for the world to prioritize AI safety in public, it's also true that, in private, concerns exist over whether rushing out products compromises this safety-first stance. This came to a head in November 2023 when OpenAI's board of directors ousted Altman, only for the company chief to return days later.

Keumars Afifi-Sabet is a freelance contributor for Tech Radar and the Technology Editor for Live Science. He has written for a variety of publications including ITPro, The Week Digital and ComputerActive. He has worked as a technology journalist for more than five years, having previously held the role of features editor with ITPro. In his previous role, he oversaw the commissioning and publishing of long form in areas including AI, cyber security, cloud computing and digital transformation.