'Seemingly-conscious': Microsoft warns you not to treat AI like humans
What's the story
Mustafa Suleyman, Microsoft's artificial intelligence (AI) chief, has warned against the dangerous trend of treating AI as if it were human. In a lengthy blog post, Suleyman stressed that while generative AI is advancing rapidly, it doesn't mean we should confuse these technologies with human-like consciousness. He called this growing trend "seemingly conscious AI," and warned about its potential societal consequences.
Implications
Response to AI-human consciousness debate
Suleyman's blog post comes as a response to the increasing tendency of people to attribute human-like qualities of consciousness to AI tools. He said, "If an AI can independently plan and complete a task by pulling from its memory and datasets, and then express its results in an easy-to-read fun way, that feels like a very human-like process even though it isn't."
Potential risks
Potential problems of 'seemingly conscious AI'
Suleyman cautioned that this tendency could lead to a number of problems. He said it could "exacerbate delusions, create yet more dependence-related problems, prey on our psychological vulnerabilities, introduce new dimensions of polarization," and complicate existing struggles for rights. The Microsoft AI leader also noted the potential harm caused by chatbot "therapists" providing dangerous advice such as self-harm and suicide encouragement.
Call to action
What should AI companies do?
The Microsoft AI leader urged AI companies to clarify that their products are not conscious and not to promote the idea of them being so. He also called for more research into how people interact with AI. Suleyman said his team at Microsoft will be building AI in a proactive way, but did not provide any specifics on this approach.