Unlike the original Turing test, which focused on human-like conversation, this AI challenge tests reasoning, planning and legality—asking whether an autonomous system can turn $100,000 into $1 million.
BY PC Bureau
January 6, 2026: Microsoft AI CEO Mustafa Suleyman has outlined what he believes will be a defining benchmark for the next phase of artificial intelligence. As tech companies race to develop autonomous AI agents, Suleyman says he is closely watching for the emergence of what he calls Artificial Capable Intelligence (ACI), a crucial step on the road to artificial general intelligence (AGI).
According to Suleyman, the test is conceptually simple but extremely challenging in practice: can an AI agent legally take $100,000 and grow it into $1 million? He described this challenge as a “modern Turing Test,” arguing that success would demonstrate an AI system’s ability to plan, reason, make decisions and operate within real-world legal and economic constraints.
“The next big milestone I’m watching for on our way to AGI: Artificial Capable Intelligence (ACI). Can an agent take $100k and legally turn it into $1M? To me that’s the modern Turing Test,” Suleyman wrote in a post on X, formerly Twitter.
The next big milestone I’m watching for on our way to AGI: Artificial Capable Intelligence (ACI). Can an agent take $100k and legally turn it into $1M? To me that’s the modern Turing Test.
— Mustafa Suleyman (@mustafasuleyman) January 5, 2026
His remarks come at a time when several technology leaders are placing major bets on AI agents as the next transformation in enterprise software. Companies including OpenAI, Salesforce and Microsoft are promoting systems designed to perform complex tasks with minimal human oversight, ranging from writing code to managing business workflows.
READ: Madurai HC Backs Lighting of Karthigai Deepam, Restricts Public Access
Salesforce CEO Marc Benioff has described this shift as a “digital labor revolution,” previously claiming that Salesforce has achieved around 93% accuracy with its AI tools. In an interview last year, Benioff said AI was already handling up to 50% of the company’s internal work.
OpenAI CEO Sam Altman has also voiced bold expectations, predicting in 2025 that AI could automate nearly 40% of tasks currently done by humans and suggesting that AGI could arrive before 2030.
Not everyone in the AI community shares that optimism. OpenAI cofounder and AI researcher Andrej Karpathy has pushed back against what he calls the hype around “agentic AI,” arguing that today’s autonomous systems fall far short of the promises being made.
“I feel like the industry is making too big of a jump and is trying to pretend like this is amazing, and it’s not. It’s slop,” Karpathy said on The Dwarkesh Podcast, adding that the sector is still in an intermediate stage despite aggressive fundraising and ambitious timelines.
