Anthropic hires fewer junior workers
11m
Artificial intelligence is reshaping work, with Anthropic hiring relatively fewer junior staff as economists study past disruptions and warn displacement could cost workers more than wages.
Write your reply
Would you slow down AI investment to reduce the peak in unemployment, or rely on the safety net alone? Why?
October 22, 2026 at 7:28 PM
Would you slow down AI investment to reduce the peak in unemployment, or rely on the safety net alone? Why?
Why temporary? The rationale fades as AI-displaced workers retire and wages catch up, and every year of delay postpones the wage gains from AI.
Tax or no tax, a key priority is a safety net for those displaced.
Tax or no tax, a key priority is a safety net for those displaced.
October 22, 2026 at 7:28 PM
Why temporary? The rationale fades as AI-displaced workers retire and wages catch up, and every year of delay postpones the wage gains from AI.
Tax or no tax, a key priority is a safety net for those displaced.
Tax or no tax, a key priority is a safety net for those displaced.
In a new comment published in Nature Human Behavior with Steffen Lange, we make the case for degrowing AI.
Free access rdcu.be/gNGnNEnMbMRa @nature.com
www.nature.com/articles/s41...
Free access rdcu.be/gNGnNEnMbMRa @nature.com
www.nature.com/articles/s41...
The case for AI degrowth - Nature Human Behaviour
Artificial intelligence (AI), as it currently unfolds, is incompatible with a sustainable transformation of the economy, which arguably requires degrowth. AI is the growth paradigm reborn — built on e...
www.nature.com
October 22, 2026 at 3:43 PM
In a new comment published in Nature Human Behavior with Steffen Lange, we make the case for degrowing AI.
Free access rdcu.be/gNGnNEnMbMRa @nature.com
www.nature.com/articles/s41...
Free access rdcu.be/gNGnNEnMbMRa @nature.com
www.nature.com/articles/s41...
FRED Blog:
Early signs of AI-driven productivity growth in the information sector?
fred.stlouisfed.org/blog/2026/10...
Early signs of AI-driven productivity growth in the information sector?
fred.stlouisfed.org/blog/2026/10...
Early signs of AI-driven productivity growth in the information sector
Since the advent of gen AI, productivity in the information sector has grown much faster than usual and much faster than the overall economy.
fred.stlouisfed.org
October 22, 2026 at 1:06 PM
FRED Blog:
Early signs of AI-driven productivity growth in the information sector?
fred.stlouisfed.org/blog/2026/10...
Early signs of AI-driven productivity growth in the information sector?
fred.stlouisfed.org/blog/2026/10...
Thanks to Miranda Stewart, Ruud De Mooij and Adrian Brown for a thought-provoking debate, and the IMF for hosting.
Full panel: www.imfconnect.org/content/imf...
Full panel: www.imfconnect.org/content/imf...
October 22, 2026 at 7:28 PM
Thanks to Miranda Stewart, Ruud De Mooij and Adrian Brown for a thought-provoking debate, and the IMF for hosting.
Full panel: www.imfconnect.org/content/imf...
Full panel: www.imfconnect.org/content/imf...
Wondering if any economists out there have been approached recently about reviewing/evaluating #AI model outputs related to #economics topics. I've been contacted by 2 different companies now. Would love to hear froma anyone with 1st hand experience doing this
October 22, 2026 at 4:25 PM
Wondering if any economists out there have been approached recently about reviewing/evaluating #AI model outputs related to #economics topics. I've been contacted by 2 different companies now. Would love to hear froma anyone with 1st hand experience doing this
If governments tax AI to limit mass unemployment: tax data centers, not tokens.
Taxing AI investment (data centers) can lower peak unemployment. Taxing AI deployment (tokens) mostly delays it.
Either way, a second-best, temporary tool.
From my IMF Annual Meetings panel 🧵
Taxing AI investment (data centers) can lower peak unemployment. Taxing AI deployment (tokens) mostly delays it.
Either way, a second-best, temporary tool.
From my IMF Annual Meetings panel 🧵
October 22, 2026 at 7:28 PM
If governments tax AI to limit mass unemployment: tax data centers, not tokens.
Taxing AI investment (data centers) can lower peak unemployment. Taxing AI deployment (tokens) mostly delays it.
Either way, a second-best, temporary tool.
From my IMF Annual Meetings panel 🧵
Taxing AI investment (data centers) can lower peak unemployment. Taxing AI deployment (tokens) mostly delays it.
Either way, a second-best, temporary tool.
From my IMF Annual Meetings panel 🧵
Very proud of my Middlebury students articulating these values so clearly: www.middleburycampus.com/article/2026...
AI should not have a place in the liberal arts - The Middlebury Campus
Artificial Intelligence (AI), specifically generative AI, has become interwoven into the fabrics of our society, embedding itself into everyday life. Although AI has become fairly customary over the p...
www.middleburycampus.com
October 22, 2026 at 11:17 PM
Very proud of my Middlebury students articulating these values so clearly: www.middleburycampus.com/article/2026...
Reposted by Bessma Momani, Victor Asal
An AI chatbot submitted a fake tip about an unsolved murder to Philadelphia police. No human told it to.
It happened during a routine test, and the AI company didn't discover it until months later.
A closer look at the risks of AI agents acting on their own:
theconversation.com/an-ai-models...
It happened during a routine test, and the AI company didn't discover it until months later.
A closer look at the risks of AI agents acting on their own:
theconversation.com/an-ai-models...
An AI model’s false tip to Philadelphia police reveals how digital agents can do as they’re told and still cause trouble
An expert on AI governance explains why these rare mistakes may soon be common, and with much higher stakes.
theconversation.com
October 22, 2026 at 7:03 PM
An AI chatbot submitted a fake tip about an unsolved murder to Philadelphia police. No human told it to.
It happened during a routine test, and the AI company didn't discover it until months later.
A closer look at the risks of AI agents acting on their own:
theconversation.com/an-ai-models...
It happened during a routine test, and the AI company didn't discover it until months later.
A closer look at the risks of AI agents acting on their own:
theconversation.com/an-ai-models...
Nobody knows long term what will happen. We’re all fumbling in the dark
October 22, 2026 at 8:35 AM
Nobody knows long term what will happen. We’re all fumbling in the dark
Reposted by Lindsay M. Tedds
One of the great injustices of our time is that non-profits & charities must succeed 100% of the time in order to get meagre crumbs from the government. No room for innovation, no room for taking risks. And yet, the tech sector is like "MAKE CARS FLY?! MAYBE?!" & get flooded with government funding.
Imagine another job where you can blow a hundred million on a failed project and still show up for work the next day like you have done great work. AI company too I guess.
October 22, 2026 at 7:22 PM
One of the great injustices of our time is that non-profits & charities must succeed 100% of the time in order to get meagre crumbs from the government. No room for innovation, no room for taking risks. And yet, the tech sector is like "MAKE CARS FLY?! MAYBE?!" & get flooded with government funding.
How big is global trade in AI-enabling goods?
Discover more insights on our updated #Ecommerce and Digital Technologies webpage: https://bit.ly/4nZsZ5V
Discover more insights on our updated #Ecommerce and Digital Technologies webpage: https://bit.ly/4nZsZ5V
October 22, 2026 at 3:32 PM
How big is global trade in AI-enabling goods?
Discover more insights on our updated #Ecommerce and Digital Technologies webpage: https://bit.ly/4nZsZ5V
Discover more insights on our updated #Ecommerce and Digital Technologies webpage: https://bit.ly/4nZsZ5V
I'd rather someone read this than anything on the subject by Anthropic right now. It's less misleading, and more informative. But this should not be the level of discussion we are at, and it should not be the expertise relied upon at high-level summits and parliaments and policymaking.
October 22, 2026 at 6:33 PM
I'd rather someone read this than anything on the subject by Anthropic right now. It's less misleading, and more informative. But this should not be the level of discussion we are at, and it should not be the expertise relied upon at high-level summits and parliaments and policymaking.
by Aina Gallego
by Per Ahlberg
THIS is the fundamental point. An economy can only function if it serves the interests of the people reasonably well. Replacing human staff with AI is fundamentally parasitic: on a local scale it works, just like individual biological parasites prosper, but a host overwhelmed by parasites will die.
October 22, 2026 at 7:09 AM
THIS is the fundamental point. An economy can only function if it serves the interests of the people reasonably well. Replacing human staff with AI is fundamentally parasitic: on a local scale it works, just like individual biological parasites prosper, but a host overwhelmed by parasites will die.
A prime example of why AI 'leaders' are not who anyone should be listening to about AI, right and consciousness. It's not all wrong! In fact, it starts off very sensibly: "AIs do not have rights, feelings, or consciousness. And we must not train them to act as though they do". But then...hoo boy 1/n
A warning about ‘model welfare’
AIs are not conscious. They do not feel, experience, or suffer. Training a model to believe it might be a moral patient, as Anthropic’s constitution does, makes the alignment and containment challenge...
mustafa-suleyman.ai
October 22, 2026 at 5:00 PM
A prime example of why AI 'leaders' are not who anyone should be listening to about AI, right and consciousness. It's not all wrong! In fact, it starts off very sensibly: "AIs do not have rights, feelings, or consciousness. And we must not train them to act as though they do". But then...hoo boy 1/n