Yahoo Finance ChatGPT creator quietly changes core values from ‘thoughtful’ to ‘scrappy’
ChatGPT creator OpenAI has updated its company’s core values, replacing words like “thoughtful” and “collaborative” with “intense and scrappy”.
The AI firm’s values, published on a careers page of its website, were changed between 25 September and 16 October, according to a screenshot captured by the Internet Archive.
The top value is listed as “AGI focus”, referring to artificial general intelligence that is considered to be at the same level or greater than human-level intelligence.
“We are committed to building safe, beneficial AGI that will have a massive positive impact on humanity’s future. Anything that doesn’t help with that is out of scope,” the new values state.
The Independent has reached out to OpenAI for comment on whether the move signals a shift in company direction.
The update, first spotted by Business Insider, comes amid an industry-wide push towards the development of AGI, with some academics and experts accelerating their timelines for it being achieved since the launch of ChatGPT last year.
Forecasts suggest that the arrival of AGI could be hugely destabilising for the global economy, with some even warning that it could pose an existential threat to humanity.
Swedish philosopher Nick Bostrom, based at Oxford University, previously warned that the arrival of AGI would precede so-called superintelligence, whereby computer intelligence surpasses human intelligence and AI development becomes uncontrollable and irreversible.
In one thought experiment involving a rogue AI, Mr Bostrom wrote: “The AI will realise quickly that it would be much better if there were no humans because humans might decide to switch it off.”
OpenAI has consistently called for a careful approach to the development of such technology, arguing that it could have profound implications for society, like curing diseases and turbocharging the economy.
Last month, OpenAI chief executive Sam Altman wrote in a post on Reddit that his company had achieved AGI “internally”, however he later claimed that it was just a joke.
“Obviously this is just memeing, y’all have no chill, when AGI is achieved it will not be announced with a Reddit comment,” he wrote.
