Microsoft AI chief says only biological beings can be conscious
Microsoft AI CEO Mustafa Suleyman speaks at the company’s 50th anniversary commemoration event at Microsoft headquarters in Redmond, Washington, on April 4, 2025.
David Ryder | Bloomberg | Getty Images
Microsoft Artificial Intelligence chief Mustafa Süleyman says that only biological beings are capable of consciousness, and that developers and researchers should stop pursuing projects that suggest otherwise.
“I don’t think this is a job that people should be doing,” Suleiman told CNBC in an interview this week where he was among the keynote speakers at the AfroTech Conference in Houston. “If you ask the wrong question, you’ll get the wrong answer. I think it’s exactly the wrong question.”
Suleiman, Microsoft’s senior executive who works on artificial intelligence, has been a leading voice in the rapidly evolving field speaking out against the possibility of apparently conscious artificial intelligence, or artificial intelligence services that can convince people they are capable of suffering.
In 2023, he co-authored the book “The Coming Wave,” which examines the risks of artificial intelligence and other emerging technologies. And in August, Suleiman wrote: article It was titled “We should build artificial intelligence for humans, not to be humans.”
This is a controversial issue as the AI complement market is growing rapidly with products from the following companies: Meta and Elon Musk’s xAI. This becomes a complex problem as the generative AI market, led by Sam Altman and OpenAI, moves towards artificial general intelligence (AGI), or AI that can perform intellectual tasks at the same level as humans’ abilities.
Altman told CNBC’s “Squawk Box” in August that AGI is “not a super useful term” and that in fact the models are advancing rapidly and we’ll be relying on them “for more and more things.”
According to Suleyman, it is particularly important to draw a clear contrast between AI becoming increasingly smarter and more capable and its ability to have human emotions.
“Our experience of physical pain is something that makes us feel very sad and terrible, but when AI experiences ‘pain’ it doesn’t feel sad,” Suleiman said. “That’s a very, very important distinction. It’s actually just creating the perception, it’s creating the apparent narrative of the experience, of itself, of consciousness, but that’s not what it’s actually experiencing. Technically you know that because we can see what the model is doing.”
In the field of artificial intelligence, there is a theory called biological naturalism, proposed by philosopher John Searle, which says that consciousness depends on the processes of a living brain.
“The reason we give rights to people today is because we don’t want to harm them, because they are suffering. They have a network of pain and they have preferences that involve avoiding pain,” Suleiman said. “These models do not have this feature. It is just a simulation.”
Solomon and others have said that the science of detecting consciousness is still in its infancy. He stopped short of saying that others should be prevented from investigating the issue and acknowledged that “different organizations have different missions.”
However, Süleyman emphasized how strongly he opposed this idea.
“They’re not conscious,” he said. “So it would be absurd to continue research investigating this question, because they are not and cannot be.”
‘Places we won’t go’
Suleiman is on a speaking tour, in part to inform the public about the risks of pursuing AI awareness.
He spoke at the Paley International Council Summit in Silicon Valley last week, ahead of the AfroTech Conference. Here is Solomon He said Microsoft won’t create chatbots for erotica; It’s a stance at odds with others in the tech industry. Altman announced in October that ChatGPT would allow adult users to engage in erotic chats, while xAI offers a risqué anime companion.
“You can basically buy these services from other companies, so we’re making decisions about which places we won’t go to,” said Suleiman at AfroTech.
Suleiman joined Microsoft in 2024 after the company paid $650 million in a licensing and acquisition deal for startup Inflection AI. Previously founded and sold DeepMind Google ten years ago for $400 million.
In a Q&A session at AfroTech, Suleiman said he decided to join Microsoft last year in part because of the company’s history, stability and broad technological reach. It was also followed by CEO Satya Nadella.
“The other thing to say is that Microsoft needs to be self-sufficient in AI,” he said on stage. “Our CEO, Satya, started this mission about 18 months ago to make sure we had the capacity internally to train our own models end-to-end with all of our data, pre-training, post-training, judgment, and deployment across products. And that was part of why I hired my team.”
Microsoft has been a major investor and cloud partner of OpenAI since 2019, and the companies have used their strengths to build large AI businesses. But the relationship has started to show signs of tension recently, with OpenAI partnering with Microsoft’s rivals such as Google and Microsoft. Seerand Microsoft is focusing more on its own AI services.
Süleyman’s concerns about consciousness resonated. In October, California Governor Gavin Newsom signed SB 243, which requires chatbots to disclose that they are artificial intelligence and to tell minors to “take a break” every three hours.
Microsoft last week announced New features for the Copilot AI service, including an AI companion named Mico and the ability to join group chats with others with Copilot. Süleyman said that Microsoft is developing services that are aware of artificial intelligence.
“It’s very simple, we always create artificial intelligence that works in the service of humans,” he said.
“There’s plenty of room for personality,” he added.
“The information is available and the models are very, very sensitive,” Suleiman said. “It is everyone’s responsibility to try to shape AI personalities with the values they want to see, use and interact with.”
Solomon highlighted a feature that Microsoft launched last week called real conversation, a style of conversation that Copilot uses designed to challenge users’ perspectives rather than flattery.
Suleiman described the actual speech as arrogant and said it had angered him recently, calling it “the biggest pile of contradictions” for warning about the dangers of artificial intelligence in his book while also accelerating its development at Microsoft.
“This was just a magical use case because in some ways I actually felt like it understood me,” Suleyman said, noting that artificial intelligence is full of contradictions.
“It’s both very impressive in some ways, but also completely magical,” he said. “And if you’re not afraid of it, you don’t really understand it. You should be afraid of it. Fear is healthy. Skepticism is necessary. We don’t need unbridled accelerationism.”
WRISTWATCH: Jefferies’ Brent Thill says Microsoft is making money from AI
