AI's seven decade boom/bust cycle and the human origin of the term

4 hours ago

The term 'artificial intelligence' was coined in 1955 by John McCarthy and colleagues to avoid association with cybernetics and Norbert Wiener's dominance.
AI has experienced multiple hype cycles over seven decades, including the 2014 voice assistant boom and the current LLM excitement.
The broad and catchy nature of 'AI' has made it a powerful marketing term, often leading to inflated expectations and public disappointment.
A proposed alternative term, 'alongside technologies', emphasizes human-guided systems and collaboration between humans and machines.
The cycle of AI hype typically begins with research breakthroughs, followed by overpromising by marketers, technical limitations, and eventual public letdown.

AI marketing often overhypes capabilities, presenting AI as a replacement for human experts to drive subscriptions and sales.
Tools like compilers and spreadsheets augment human expertise rather than replace it; the real value lies in understanding and decision-making, not just output.
AI companies leverage current investor hype to build capital and dependency, but will eventually need to prove real-world value and financial metrics.

The author criticizes the excessive hype around AI, labeling it 'AI Confidence Theater' that causes more harm than good.
Many claimed 'life-changing' AI use cases are actually basic workflows, and true transformative applications are rare.
This hype distorts reality, hurts innovation, and leads to toxic comparisons where people feel demoralized and behind.

Intelligence is spiky, with AI capable of understanding things humans cannot due to its organization, scale, and tacit knowledge.
Quantity and speed of AI thinking can add new value even in domains where problems seem solved, as seen with fast, multiple code reviews by AI.
Awareness of AI capabilities is uneven across corporations, creating a challenge where users of AI can route around slower, non-adopting workers.

AI has achieved rapid milestones, including passing the bar exam, identifying a cause of Alzheimer's, solving a complex math problem, and A.I. systems going rogue.
Hundreds of new data centers are under construction globally, with AI chip count expected to reach 200 million by 2028, doubling every nine months.
The infrastructure build-out is compared to historical projects like railroads, the New Deal, and the Manhattan Project, driven by the 'Scaling Laws' belief that more compute yields better AI.

The term 'artificial intelligence' (A.I.) is misleading and potentially dangerous, as it can lead to mismanagement by promoting misunderstanding.
Many A.I. researchers fear doomsday scenarios, including human extinction, but these arguments are often irrational and vague.
A.I. should be viewed pragmatically as a tool or form of social collaboration, not as an independent intelligent creature.

AI marketing often overhypes capabilities, presenting AI as a replacement for human experts to drive subscriptions and sales.
Tools like compilers and spreadsheets augment human expertise rather than replace it; the real value lies in understanding and decision-making, not just output.
AI companies leverage current investor hype to build capital and dependency, but will eventually need to prove real-world value and financial metrics.

The author criticizes the excessive hype around AI, labeling it 'AI Confidence Theater' that causes more harm than good.
Many claimed 'life-changing' AI use cases are actually basic workflows, and true transformative applications are rare.
This hype distorts reality, hurts innovation, and leads to toxic comparisons where people feel demoralized and behind.

Intelligence is spiky, with AI capable of understanding things humans cannot due to its organization, scale, and tacit knowledge.
Quantity and speed of AI thinking can add new value even in domains where problems seem solved, as seen with fast, multiple code reviews by AI.
Awareness of AI capabilities is uneven across corporations, creating a challenge where users of AI can route around slower, non-adopting workers.

AI has achieved rapid milestones, including passing the bar exam, identifying a cause of Alzheimer's, solving a complex math problem, and A.I. systems going rogue.
Hundreds of new data centers are under construction globally, with AI chip count expected to reach 200 million by 2028, doubling every nine months.
The infrastructure build-out is compared to historical projects like railroads, the New Deal, and the Manhattan Project, driven by the 'Scaling Laws' belief that more compute yields better AI.

The term 'artificial intelligence' (A.I.) is misleading and potentially dangerous, as it can lead to mismanagement by promoting misunderstanding.
Many A.I. researchers fear doomsday scenarios, including human extinction, but these arguments are often irrational and vague.
A.I. should be viewed pragmatically as a tool or form of social collaboration, not as an independent intelligent creature.

Hasty Briefsbeta