AI Progress Chart Sparks Debate Over Meaning and Mechanism
In brief
- A controversial chart claiming to show AI progress has sparked intense discussion in the tech community.
- The graph, often referred to as the "METR graph," plots what its creator calls "time horizons" on the Y-axis over time.
- While the exact definition of "time horizons" remains unclear-some speculate it relates to data rather than actual time-the chart shows a steady upward trend from 2018 to 2025.
- The debate centers on whether this graph is meaningful.
- Critics argue that without a clear explanation of what "time horizons" measure, the chart offers little insight into AI progress.
- Proponents counter that it provides at least some quantitative basis for evaluating AI development, which is rare in the field.
- The discussion highlights a broader issue in AI benchmarking: the lack of standardized metrics and deeper mechanisms to explain trends.
- As the debate continues, experts suggest focusing on how changes in data availability or model architectures might impact future progress.
- This could help refine the chart's usefulness and provide more reliable insights into AI development.
Terms in this brief
- METR graph
- A chart that plots 'time horizons' over time from 2018 to 2025, sparking debate on AI progress. The exact meaning of 'time horizons' is unclear, but it may relate to data rather than actual time.
Read full story at LessWrong →
More briefs
Gen Z Uses AI More Than Anyone
Gen Z graduates booed commencement speakers who mentioned artificial intelligence. They use AI more than anyone for productivity and brainstorming. Gen Z knows AI fluency is needed for the job market. They use AI tools to get work done. About 70 percent of Gen Z students use AI to help with homework. They want to learn how to work with AI tools. Gen Z will need AI skills to succeed in the future.
Debating the Future of AI: Is It as Normal as We Think?
Arvind Narayanan and Sayash Kapoor's essay, "AI as Normal Technology," has sparked a heated debate. The authors argue that AI should be treated like any other technology, with manageable risks and gradual societal changes. However, critics like Olivia Helens and Scott Alexander challenge this view, pointing out the unique dangers of superintelligent AI. They highlight biosecurity risks and compare AI's potential impact to historical events like the chess revolution, suggesting AI could bring abrupt and far-reaching consequences. The essay's main flaws include downplaying the speed of AI adoption and underestimating its societal impact. Narayanan and Kapoor suggest a slow transformation, but critics argue that AI's rapid development could lead to sudden shifts in industries and labor markets. For instance, Olivia Helens emphasizes biosecurity as a critical risk, while Scott Alexander draws parallels from chess history to illustrate how quickly AI could disrupt systems. This debate underscores the need for cautious optimism. While AI offers immense potential, its risks require urgent attention. As experts continue to discuss these issues, the focus should be on developing robust governance frameworks and preparing for the unexpected. The future of AI remains uncertain, but one thing is clear: its impact will be profound.
Gen Z Boos AI at Graduation Ceremonies
Students at graduation ceremonies are booing speakers who praise AI. This happened to former Google CEO Eric Schmidt and Big Machine Records CEO Scott Borchetta. They were booed when they talked about AI at the ceremonies. A survey found that Gen Z's excitement about AI dropped 14% in a year. Instead, they feel angry about AI. 44% of Gen Z workers said they resisted their company's AI plans. They do not want AI to take their jobs. Many young people are worried about AI taking their jobs. Some have been looking for work for months. Companies are using AI as a reason for layoffs. Now students are speaking out against AI at graduation ceremonies. The backlash against AI will likely continue.
AI Writing Scandal Hits Literary Prize
A short story that won a prestigious literary award may have been written by AI. The story was named the winning entry for the Commonwealth prize from the Caribbean. The story's author is reportedly a 61-year-old from Trinidad and Tobago with few publications to his name. An AI detector said the work was AI-generated, citing obvious markers of AI writing. The incident has sparked a conversation about whether artists are passing off AI-generated work as their own. A decision on the prize winner's authorship may never be made. The debate over AI writing will continue in the future.
Nobel Winner Uses AI to Write Novel
Nobel Prize winner Olga Tokarczuk used AI to help write her latest novel. She asked the AI for song titles her characters would listen to and for ideas to develop her story. Tokarczuk believes AI is an advantage for literary fiction, despite its potential for errors. She thinks writers will work closely with AI tools because their creative process is based on broad associations of facts. Tokarczuk is now working on her final project, and she expects AI to play a big role in its creation. She will likely publish her last book soon.