Robert Scoble

Says Tesla is automating the car, Optimus physical work, and the next thing to automate is the company itself; he expects the first $1B company with zero HR employees and says every big shift starts with boring work nobody wants. The rest of the post promotes Warp’s AI HR agents and reads as a product plug; only his framing is used. Full post text inspected via the X API.

Says the government has put “USA’s best AI company” on pause while China has its full force on the accelerator, so America has its hands tied behind its back; still, he is optimistic because the country has often chosen the hard road. He does not name the company or the government action. Full post text inspected via the X API.

Quoting a list of AI startups that are hiring (someone else’s post), says big AI funding turns into jobs and asks whether universities are giving students the skills to get them and why students are not hearing that AI is where their future jobs are. Full post text inspected via X’s public embed endpoint.

Says he trusts his life to Tesla’s self-driving cars and hopes all robotics companies succeed, but in consumer research he keeps hearing the belief that the industry is building a “Terminator” robot that will eventually kill people, while autonomous-car companies are building the opposite belief that their robots save lives. Argues that trust will decide which robot companies people buy from. About public perception, not a risk estimate. Full post text inspected via the X API.

A one-line reply to a post saying AI change is still in its “slowly” phase before the “all at once” phase (someone else’s words). Shows his recurring theme that people underestimate exponential change. Full post text inspected via X’s public embed endpoint.

Quoting a chart claiming AI task-length doubling time has fallen to two months (someone else’s claim), says many people wrongly said LLMs had tapped out; the real problem is that better AI draws more users, causing errors at Anthropic, and maybe people should run models locally on powerful Macs. Full post text inspected via the X API.

Argues that if every state but one banned AI, that state would outproduce the others 10 to 1 within a decade, possibly far more, as a society run on robots beats one that is not. Favors a loose federal approach to encourage innovation and keep up with China, expects America to “do the stupid thing”, and says forcing one hand behind our back puts the country at severe risk. Full post text inspected via the X API.

Introducing a startup that pays people to share photos and videos to train robots and agents, says many old jobs will be automated in the next few years, but new kinds of jobs like this will put millions to work in a world with millions and eventually billions of robots and AI agents doing old jobs; we need many more such job-creating companies. Full post text inspected via the X API.

Fifteen predictions for 2035: most people will not drive, robots will be in businesses and homes, brain-computer interfaces will merge humans with AIs, people will have dozens of virtual beings, corporations will be hybrids of humans and AIs, AI will make us more productive and “even, happier”, and humans will still be needed to work, on new things that are hard to imagine. Ends by saying he already lives this way, talking with Grok all day. Older context; full post text inspected via the X API.