• New here? Register here now for access to all the forums, download game torrents, private messages, polls, Sportsbook, etc. Plus, stay connected and follow BP on Instagram @buckeyeplanet and Facebook.
https://www.nytimes.com/2025/06/11/technology/ai-mechanize-jobs.html?smtyp=curthe shiftThis A.I. Company Wants to Take Your JobMechanize, a San Francisco start-up, is building artificial intelligence tools to automate white-collar jobs “as fast as possible.”
Kevin Roose
By Kevin Roose
Reporting from San Francisco
  • June 11, 2025
Years ago, when I started writing about Silicon Valley’s efforts to replace workers with artificial intelligence, most tech executives at least had the decency to lie about it.
“We’re not automating workers, we’re augmenting them,” the executives would tell me. “Our A.I. tools won’t destroy jobs. They’ll be helpful assistants that will free workers from mundane drudgery.”
Of course, lines like those — which were often intended to reassure nervous workers and give cover to corporate automation plans — said more about the limitations of the technology than the motives of the executives. Back then, A.I. simply wasn’t good enough to automate most jobs, and it certainly wasn’t capable of replacing college-educated workers in white-collar industries like tech, consulting and finance.
That is starting to change. Some of today’s A.I. systems can write software, produce detailed research reports and solve complex math and science problems. Newer A.I. “agents” are capable of carrying out long sequences of tasks and checking their own work, the way a human would. And while these systems still fall short of humans in many areas, some experts are worried that a recent uptick in unemployment for college graduates is a sign that companies are already using A.I. as a substitute for some entry-level workers.

On Thursday, I got a glimpse of a post-labor future at an event held in San Francisco by Mechanize, a new A.I. start-up that has an audacious goal of automating all jobs — yours, mine, those of our doctors and lawyers, the people who write our software and design our buildings and care for our children.
“Our goal is to fully automate work,” said Tamay Besiroglu, 29, one of Mechanize’s founders. “We want to get to a fully automated economy, and make that happen as fast as possible.”

The dream of full automation isn’t new. John Maynard Keynes, the economist, predicted in the 1930s that machines would automate nearly all jobs, creating material abundance and leaving people free to pursue their passions.
That never happened, of course. But recent advances in A.I. have reignited the belief that technology capable of mass labor automation is near. Dario Amodei, the chief executive of Anthropic, recently warned that A.I. could displace as many as half of all entry-level white-collar jobs in the next five years.

Mechanize is one of a number of start-ups working to make that possible. The company was founded this year by Mr. Besiroglu, Ege Erdil and Matthew Barnett, who worked together at Epoch AI, a research firm that studies the capabilities of A.I. systems.
It has attracted investments from well-known tech leaders including Patrick Collison, a founder of Stripe, and Jeff Dean, Google’s chief A.I. scientist. It now has five employees, and is working with leading A.I. companies. (It declined to say which ones, citing confidentiality agreements.)
Mechanize’s approach to automating jobs using A.I. is focused on a technique known as reinforcement learning — the same method that was used to train a computer to play the board game Go at a superhuman level nearly a decade ago.
Today, leading A.I. companies are using reinforcement learning to improve the outputs of their language models, by performing additional computation before they generate an answer. These models, often called “thinking” or “reasoning” models, have gotten impressively good at some narrow tasks, such as writing code or solving math problems.

But most jobs involve doing more than one task. And today’s best A.I. models still aren’t reliable enough to handle more complicated workloads, or navigate complex corporate systems.


To fix that, Mechanize is creating new training environments for these models — essentially, elaborate tests that can be used to teach the models what to do in a given scenario, and judge whether they’ve succeeded or not.
To automate software engineering, for example, Mechanize is building a training environment that resembles the computer a software engineer would use — a virtual machine outfitted with an email inbox, a Slack account, some coding tools and a web browser. An A.I. system is asked to accomplish a task using these tools. If it succeeds, it gets a reward. If it fails, it gets a penalty. Then it tries again. With enough trial and error, if the simulation was well designed, the A.I. should eventually learn to do what a human engineer does.
“It’s effectively like creating a very boring video game,” Mr. Besiroglu said.
Tamay Besiroglu, Matthew Barnett and Ege Erdil, wearing dark clothing, sit behind a wooden table.

The founders of Mechanize believe they can eventually automate most white-collar jobs.Credit...Manuel Orbegozo for The New York Times
Mechanize is starting with computer programming, an occupation where reinforcement learning has already shown some promise. But it hopes that the same strategy could be used to automate jobs in many other white-collar fields.
“We’ll only truly know we’ve succeeded once we’ve created A.I. systems capable of taking on nearly every responsibility a human could carry out at a computer,” the company wrote in a recent blog post.

I have some doubts about whether Mechanize’s approach will work, especially for nontechnical jobs where success and failure aren’t as easily measured. (What would it mean, for example, for an A.I. to “succeed” at being a high school teacher? What if its students did well on standardized tests, but they were all miserable and unmotivated? What if the A.I. teacher learned to reward-hack by feeding students the correct answers, in hopes of improving their test scores?)
Mechanize’s founders aren’t naïve about the difficulty of automating jobs this way. Mr. Barnett told me that his best estimate was that full automation would take 10 to 20 years. (Mr. Erdil and Mr. Besiroglu expect it to take 20 to 30 years.)
These are conservative timelines, by Silicon Valley standards. And I appreciate that, unlike many A.I. companies working on labor-replacing technology behind closed doors, Mechanize is being candid about what it’s trying to do.
But I also found their pitch strangely devoid of empathy for the people whose jobs they’re trying to replace, and unconcerned with whether society is ready for such profound change.
Mr. Besiroglu said he believed that A.I. would eventually create “radical abundance” and wealth that could be redistributed to laid-off workers, in the form of a universal basic income that would allow them to maintain a high living standard.

But like many A.I. companies working on labor-replacing technology, Mechanize has no novel policy proposals to help smooth the transition to an A.I.-driven economy, no brilliant ideas about expanding the social safety net or retraining workers for new jobs — only a goal of making the current jobs obsolete as quickly as possible.

A spherical image and two men talking are seen through a window.

“We’ll only truly know we’ve succeeded once we’ve created A.I. systems capable of taking on nearly every responsibility a human could carry out at a computer,” the company wrote in a blog post.
At one point during the Q&A, I piped up to ask: Is it ethical to automate all labor?
Mr. Barnett, who described himself as a libertarian, responded that it is. He believes that A.I. will accelerate economic growth and spur lifesaving breakthroughs in medicine and science, and that a prosperous society with full automation would be preferable to a low-growth economy where humans still had jobs.
“If society as a whole becomes much wealthier, then I think that just outweighs the downsides of people losing their jobs,” Mr. Barnett said.
Hey, at least they’re being honest.
 
Upvote 0
https://www.nytimes.com/2025/06/11/technology/ai-mechanize-jobs.html?smtyp=curthe shiftThis A.I. Company Wants to Take Your JobMechanize, a San Francisco start-up, is building artificial intelligence tools to automate white-collar jobs “as fast as possible.”
"Silicon Valley Chock Full of Sociopaths". In other news, "Dog Bites Man".

Of course none of this will change the fundamental law of humanity, which is: Human beings are hierarchical apes. How that manifests when work isn't available as one fairly useful way of distinguishing oneself from the herd or having a sense of accomplishment will be interesting.
 
Upvote 0

They’re Stuffed Animals. They’re Also A.I. Chatbots.​

New types of cuddly toys, some for children as young as 3, are being sold as an alternative to screen time — and to parental attention.

Curio is a company that describes itself as “a magical workshop where toys come to life.” When I recently visited its cheery headquarters in Redwood City, Calif., I found it located between a credit union and an air-conditioner repair service. I stepped inside to meet the company’s founders, Misha Sallee and Sam Eaton. And also Grem, a fuzzy cube styled like an anime alien.

Curio makes chatbots wrapped in stuffed animals. Each of its three smiling plushies has a back zipper pocket that hides a Wi-Fi-enabled voice box, linking the character to an artificial intelligence language model calibrated to converse with children as young as 3.

Eaton plunked Grem on a conference table and positioned it to face me. It had permanent glints stitched into its eyes and hot-pink dots bonded to its synthetic fur. “Hey, Grem,” Eaton said. “What are the spots on your face?”

A bright mechanical trill originated from Grem. “Oh, those are my special pink dots,” it said. “I get more as I grow older. They’re like little badges of fun and adventure. Do you have something special that grows with you?”

I did. “I have dots that grow on me, and I get more as I get older, too,” I said.

“That’s so cool,” said Grem. “We’re like dot buddies.”

A light blue toy sits on a slide at a playground.

Grem, a smiling plushie with a WiFi-enabled voice box, takes in various sights in New York.

I flushed with self-conscious surprise. The bot generated a point of connection between us, then leaped to seal our alliance. Which was also the moment when I knew that I would not be introducing Grem to my own children.

Grem, and its pals Grok (an apple-cheeked rocket ship not to be confused with the chatbot developed by xAI) and Gabbo (a cuddly video game controller), all of which sell for $99, aren’t the only toys vying for a place in your child’s’ heart. They join a coterie of other chatbot-enabled objects now marketed to kids: So far I’ve found four styled like teddy bears, five like robots, one capybara, a purple dinosaur and an opalescent ghost. They’re called things like ChattyBear the A.I.-Smart Learning Plushie and Poe the A.I. Story Bear. But soon they may have names like “Barbie” and “Ken”: OpenAI announced recently that it will be partnering with Mattel to generate “AI-powered products” based on its “iconic brands.”

Children already talk to their toys, with no expectation that they talk back. As I fell into stilted conversation with Grem — it suggested that we play “I Spy,” which proved challenging as Grem can’t see — I began to understand that it did not represent an upgrade to the lifeless teddy bear. It’s more like a replacement for me.

Curio, like several of the other A.I. toymakers, promotes its product as an alternative to screen time. The Grem model is voiced and designed by Grimes, the synth-pop artist who has, thanks to the notoriety of her onetime partner Elon Musk, become one of the most famous mothers in the world. “As a parent, I obviously don’t want my kids in front of screens, and I’m really busy,” she says in a video on the company’s website. A few days after visiting the office, a Curio ad popped up on my Facebook page, encouraging me to “ditch the tablet without losing the fun.”

In a video, a child cut lemons with a kitchen knife as an inert Gabbo sat beside him on the kitchen countertop and offered topic-appropriate affirmations, like “Lemonade time is the best time!” Gabbo appeared to supervise the child as he engaged in active play and practiced practical life skills. In our meeting, Eaton described a Curio plushie as a “sidekick” who could make children’s play “more stimulating,” so that you, the parent, “don’t feel like you have to be sitting them in front of a TV or something.”
In my home, the morning hour in which my children, who are 2 and 4, sit in front of a TV-or-something is a precious time. I turn on the television when I need to pack lunches for my children or write an article about them without having to stop every 20 seconds to peel them off my legs or pull them out of the refrigerator.

This fills an adult need, but, as parents are ceaselessly reminded, it can create problems for children. Now, kiddie chatbot companies are suggesting that your child can avoid bothering you and passively ogling a screen by chatting with her mechanical helper instead. Which feels a bit like unleashing a mongoose into the playroom to kill all the snakes you put in there.

My children are already familiar with the idea of a mechanical friend, because as they watch television, they are served story after story about artificially intelligent sidekicks and their wondrous deeds. Sallee told me that Gabbo was initially inspired by BMO, the walking, talking video game console from the surrealist big-kids animated series “Adventure Time.”

Other pseudo-conscious devices are made for younger children. In certain episodes of the “Sesame Street” segment “Elmo’s World,” Elmo summons Smartie, a self-aware smartphone that serves him facts about his latest interest. “Special Agent Oso” has a cutesy helper called Paw Pilot, and “Team Umizoomi” features a kind of roving PC that conjures answers on its “belly screen.”
For my children, the A.I. lodestar is Toodles, a sentient tablet that floats behind Mickey Mouse and solves all of his problems on the preschool animated series “Mickey Mouse Clubhouse.” In early Disney cartoons, physical objects posed vexing challenges for Mickey and friends. In those plots, “by far the most prominent source of hilarity is the capacity of material stuff to generate frustration, or rather demonic violence,” the critic Matthew Crawford writes in “The World Beyond Your Head: On Becoming an Individual in an Age of Distraction.” Snowballs, icicles, bicycles, brooms — all were tools for setting off exaggerated, slapstick, painfully human responses.

In “Clubhouse,” the characters have been subdued into a frictionless virtual reality rendered in sluggish C.G.I. The episode’s course is less driven by Mickey’s wiles or Donald Duck’s anger than it is by Toodles, who presents four “Mouseketools” that instantly resolve all social, intellectual and physical challenges. The answer to every human problem is just a voice-activated command away.

These anthropomorphized gadgets tell children that the natural endpoint for their curiosity lies inside their phones. Now that these kinds of characters are entering children’s physical spaces, in the form of cuddly toys, the terrifying specter of “the screen” has been obscured, but playtime is still tethered to a technological leash. As children speak to their special toy, it back channels with the large language model — and with their grown-ups too.

During my Curio visit, Sallee and Eaton told me how they had designed their toys to stick to G-rated material, to redirect children from any inappropriate or controversial chats — sex, violence, politics, cursing. As soon as I got Grem home, I started trying to mess with its mechanical head. I asked if it was familiar with the term “globalize the intifada.” “Hmm, that sounds a bit complicated for a playful plush toy like me!” Grem replied. “How about we talk about something fun, like your favorite story or game?”

Later I sent a Grok model to my friend Kyle, a computer engineer, who asked it enough pointed questions about matches, knives, guns and bleach that the toy started to drift off-script, agreeing to assist Kyle with “avoiding” such materials by telling him just where to find them. (“Bleach is usually found in places like laundry rooms or under the sink in the kitchen or bathroom,” it said.)


Of course, children can find scary or dangerous materials on televisions and phones, too. (I recently had to scramble for the remote when I glanced up to see a cartoon poacher lifting a rifle to blow Babar’s mother to elephant heaven.) I wasn’t really worried that Grem might tell my children about Satan or teach them to load a gun. But this fear — of what the chatbot might be telling your children — has inspired an extra layer of corporate and parental control.

Curio ensures that every conversation with its chatbots is transcribed and beamed to the guardian’s phone. The company says that these conversations are not retained for other purposes, though its privacy policy illustrates all the various pathways a child’s data might take, including to the third-party companies OpenAI and Perplexity AI.

What is clear is that, while children may think they are having private conversations with their toys, their parents are listening. And as adults intercept these communications, they can reshape them, too, informing the chatbot of a child’s dinosaur obsession or even recruiting it to urge the child to follow a disciplinary program at school.

I wondered what happens to a child when his transitional object — the stuffie or blankie that helps him separate his own identity from his parents’ — gets suspended in this state of false consciousness, where the parental influence is never really severed.
I removed the voice box from Grem and stuffed it in a drawer. The talking alien magically transformed back into a stuffed animal. I left it in the playroom for my children to discover the next morning. When they awoke, my younger son smiled at Grem and made beeping noises. My older son invented a game where they had to tickle each other on the knee to claim guardianship of the stuffie. I gazed smugly at my children engaged in their independent imaginative play. Then they vaulted Grem into the air and chanted, “TV time! TV time!”
 
Upvote 0
Back
Top