Artificial intelligence has burrowed its way into every aspect of the modern life, from intelligent vacuum cleaners and driverless cars to advanced techniques for diagnosing diseases.
Its promoters reckon it is revolutionising human experience, but critics stress that the technology risks putting machines in charge of life-changing decisions.
Regulators in Europe and North America are worried.
The European Union is likely to pass legislation next year - the AI Act - aimed at reining in the age of the algorithm.
The United States has recently published a blueprint for an AI Bill of Rights, and Canada is mulling legislation.
China is using biometric data, facial recognition and other technology to build a powerful system of control that is a big topic in the debates.
Gry Hasselbalch, a Danish academic who advises the EU on the controversial technology, said that the West was in danger of creating totalitarian infrastructures. "I see that as a huge threat, no matter what the benefits," she told AFP.
Before regulators can act, they have to deal with the daunting task of defining what AI actually is.
Suresh Venkatasubramanian of Brown University, who co-authored the AI Bill of Rights, said that trying to define AI was a mug's game. Any technology that affects people's rights should be within the scope of the bill, he tweeted.
The EU is taking the more tortuous route of trying to define the sprawling field.
Its draft law lists the kinds of approaches that are defined as AI, and it includes pretty much any computer system that involves automation.
The changing use of the term AI has resulted in a problem.
It described attempts to create machines that simulate human thinking for decades.
In the early 2000s, funding for this research - known as symbolic AI - was largely dried up.
The rise of Silicon Valley titans saw AI reborn as a catch-all label for their number-crunching programs and algorithms they generated.
This automation allowed them to target users with ads and content, helping them to make hundreds of billions of dollars.
Meredith Whittaker, a former Google worker who co-founded New York University's AI Now Institute, told AFP that AI was a way for them to make more use of the surveillance data and to mystify what was happening.
The EU and the US agree that any definition of AI needs to be as broad as possible.
Both Western powerhouses have gone their separate ways from that point onwards.
The EU's AI Act is more than 100 pages long.
The complete prohibition of certain high-risk technologies is one of its most eye-catching proposals, as well as the use of biometric surveillance tools in China.
It also limits the use of AI tools by migration officials, police and judges.
Hasselbach said some technologies were simply too challenging to fundamental rights. The AI Bill of Rights, on the other hand, is a brief set of principles framed in an aspirational language, with exhortations like you should be protected from unsafe or ineffective systems. The bill was issued by the White House and relies on existing law.
Experts think there will be no dedicated AI legislation in the United States until 2024, because Congress is deadlocked.