Former OpenAI employee: Company is building the ‘Titanic of AI’

Former OpenAI employee William Saunders shared a dire warning about his previous company, ChatGPT’s creator. 

On July 3, he said in a YouTube podcast that CEO Sam Altman was building the “Titanic of AI.” Like the disastrous ship, OpenAI is allegedly creating artificial intelligence without ensuring customer safety. 

READ: Former OpenAI researcher shares the next 10 years of AI

That is why Saunders believes the global AI brand needs to build more “lifeboats” for its technology. OpenAI is changing the world, but it must address the potential risks. 

What did the former OpenAI employee say?

William Saunders was part of OpenAI’s Superalignment team, the group of experts tasked with preventing the potential negatives of AI development. 

He and other employees stepped down after seeing the company’s trajectory, akin to the Titanic’s construction. 

White Star Line manufactured the iconic vessel that crashed into an iceberg, leading many to their deaths. “During my three years at OpenAI, I would sometimes ask myself a question,” Saunders said in the Alex Kantrowitz podcast.

“Was the path that OpenAI was on more like the Apollo program or more like the Titanic?”

He said the Apollo 13 space mission had numerous measures to ensure its crew returned to Earth safely. On the other hand, White Star Line built the Titanic to make cruise liners bigger than their competitors.

“There weren’t enough lifeboats for everyone. So when disaster struck, a lot of people died,” the previous AI safety officer explained.

Similarly, OpenAI’s “priorities are more like a product company.” It wanted “newer, shinier products,” while relying too heavily on current safety measures. 

The former OpenAI employee admitted that AI is becoming a “wildly transformative” force worldwide. However, he believes “they have a duty to put in the work to prepare for that.”

Fellow AI experts are developing safer alternatives to ChatGPT. For example, OpenAI co-founder Ilya Sutskever recently founded a new AI firm called Safe Superintelligence. 

Previous OpenAI experts formed the company Anthropic and built ChatGPT rival Claude. Learn more about this AI alternative in this other Inquirer Tech article.

Read more...