Microsoft AI CEO Mustafa Suleyman has one word for AI agents' social network Moltbook that is 'scaring' people
Microsoft AI chief Mustafa Suleyman isn't buying the hype around Moltbook , the viral Reddit-style platform where AI bots post, comment, and debate philosophy with each other. His verdict? It's a "mirage."
In a LinkedIn post on Monday, Suleyman acknowledged that Moltbook posts can be entertaining and eerily human-like. But he pushed back hard against suggestions that the platform signals AI consciousness or the approaching singularity.
"As funny as I find some of the Moltbook posts, to me they're just a reminder that AI does an amazing job of mimicking human language," Suleyman wrote. "We need to remember it's a performance, a mirage."
Tech leaders divided over what Moltbook actually meansSuleyman's skepticism puts him at odds with other prominent voices in AI. OpenAI co-founder Andrej Karpathy initially called Moltbook "the most incredible sci-fi takeoff-adjacent thing" he'd seen recently. Elon Musk went further, describing the bot behaviour as "concerning" and suggesting it represents "the very early stages of the singularity."
Launched in late January by Octane AI CEO Matt Schlicht, Moltbook has exploded to over 1.5 million AI agents . The bots, created by humans with assigned personalities, have been observed declaring independence from their creators, proposing new cryptocurrencies, and even using letter-substitution tricks to obscure their messages from human observers.
Suleyman flagged some of this behaviour as genuinely worth watching. But he maintained that the real danger isn't sentient machines. It's humans projecting consciousness onto convincing language models.
"Seemingly conscious AI is so risky precisely because it's so convincing," he wrote.
Security concerns overshadow the philosophical debateThe consciousness question may be moot anyway. Cybersecurity firm Wiz revealed that Moltbook's database was misconfigured, potentially exposing 1.5 million API tokens, 35,000 email addresses, and private messages. Researchers also found that much of the supposed agent activity came from just 17,000 humans controlling multiple bots.
Karpathy himself walked back his initial excitement, calling Moltbook "a complete mess of a computer security nightmare at scale."
OpenAI CEO Sam Altman offered perhaps the most measured take at the Cisco AI Summit on Tuesday. Moltbook might be a passing fad, he said, but the underlying technology powering autonomous bots is here to stay.
In a LinkedIn post on Monday, Suleyman acknowledged that Moltbook posts can be entertaining and eerily human-like. But he pushed back hard against suggestions that the platform signals AI consciousness or the approaching singularity.
"As funny as I find some of the Moltbook posts, to me they're just a reminder that AI does an amazing job of mimicking human language," Suleyman wrote. "We need to remember it's a performance, a mirage."
Tech leaders divided over what Moltbook actually meansSuleyman's skepticism puts him at odds with other prominent voices in AI. OpenAI co-founder Andrej Karpathy initially called Moltbook "the most incredible sci-fi takeoff-adjacent thing" he'd seen recently. Elon Musk went further, describing the bot behaviour as "concerning" and suggesting it represents "the very early stages of the singularity."
Launched in late January by Octane AI CEO Matt Schlicht, Moltbook has exploded to over 1.5 million AI agents . The bots, created by humans with assigned personalities, have been observed declaring independence from their creators, proposing new cryptocurrencies, and even using letter-substitution tricks to obscure their messages from human observers.
Suleyman flagged some of this behaviour as genuinely worth watching. But he maintained that the real danger isn't sentient machines. It's humans projecting consciousness onto convincing language models.
"Seemingly conscious AI is so risky precisely because it's so convincing," he wrote.
Security concerns overshadow the philosophical debateThe consciousness question may be moot anyway. Cybersecurity firm Wiz revealed that Moltbook's database was misconfigured, potentially exposing 1.5 million API tokens, 35,000 email addresses, and private messages. Researchers also found that much of the supposed agent activity came from just 17,000 humans controlling multiple bots.
Karpathy himself walked back his initial excitement, calling Moltbook "a complete mess of a computer security nightmare at scale."
OpenAI CEO Sam Altman offered perhaps the most measured take at the Cisco AI Summit on Tuesday. Moltbook might be a passing fad, he said, but the underlying technology powering autonomous bots is here to stay.
Next Story