Microsoft Executive Describes Superintelligent AI as a 'Counter Goal'

SadaNews - While most of Silicon Valley races to achieve superintelligent AI that surpasses human capabilities, the head of the AI department at Microsoft is trying to rein in this ambition.

Mustafa Suleyman said in an episode of the "Silicon Valley Girl" podcast aired on Saturday that the idea of superintelligent AI should not only be avoided but should also be considered a "counter goal".

Suleyman added that superintelligent AI - that is, AI capable of thinking far beyond human capabilities - "does not seem like a positive vision for the future", according to a report by "Business Insider".

He continued: "It will be very difficult to contain something like this or align it with our values".

Suleyman, who co-founded DeepMind - currently part of Google - before moving to Microsoft, noted that his team is trying to develop human-centric superintelligence, intelligence that supports human interests.

He also added that granting AI any form of awareness or moral status is a mistake, saying: "These entities do not suffer. They do not feel pain... They merely mimic high-quality conversation".

Suleyman's statements come at a time when some industry leaders are discussing the development of superintelligent AI, with some suggesting it could arrive in this decade.

The CEO of the AI company "OpenAI", Sam Altman, described general AI - intelligence that can think like humans - as the company's core mission. Altman added earlier this year that "OpenAI" has already begun looking beyond general AI towards superintelligence.