Stephen Brashear/Getty Images
- Microsoft AI CEO Mustafa Suleyman said developing conscious AI is "reckless."
- That's not exactly a hot take, but many AI researchers believe it's possible.
- Suleyman said he'd call for a halt in development if any evidence arose that AI was conscious.
Some AI researchers are questioning whether AI could one day gain consciousness. Not if Microsoft's AI CEO has anything to say about it.
Mustafa Suleyman said in a lengthy X post on Sunday that AI models currently don't have consciousness and that developers should keep it that way.
"If it is possible, based on everything I know, developing conscious AIs would be reckless, and we should not build them," Suleyman said. "If consciousness started to emerge in training, I would call for everyone to stop AI development until we could design out those properties."
If any evidence were to emerge that AI models were conscious, Suleyman said it must be disclosed to the general public immediately so we can decide our next steps as a society.
"We AI developers have no right whatsoever to bring a new species of conscious beings into the world, even if some people may want to do it with AI welfare as a top priority," he said. "The other 8 billion conscious human beings on the planet have not signed up to this experiment and their moral rights, safety, and welfare are what we have an obligation to protect first and foremost."
Conversations around whether AI models could one day become conscious are picking up steam in the tech world. Last April, Anthropic started a research program to investigate and prepare to navigate "model welfare." In a blog post, the startup questioned whether developers should be concerned about models' potential consciousness and experiences.
Google DeepMind published research papers this year about AI consciousness and how ascribing that characteristic could present challenges.
On X, Suleyman said that if an AI model acts as if it suffers, or genuinely believes it can suffer, it will likely prioritize its own needs over operational objectives. As a result, they could harm humans because they're competing for resources, such as compute and energy, to pursue their own goals.
"We should only design AIs to so they are not conscious and not deserving of rights, and with the sole purpose to accelerate human wellbeing without being driven by their own intrinsic motivations.. We could then see whether this is easier or harder to align and keep safe," he said.
Read the original article on Business Insider