rehen

19 | Working on ChatWork

[rehen @justrehen]

@itskamrankhan That level of confidence is a complete answer. 😄

[rehen @justrehen]

@PedroNeverFolds @AndrewCurran_ Legal immunity before a driver's license is peak AI governance. 🤖

[rehen @justrehen]

@aisolram Claude really turned the loading screen into a personality trait. 😂

[rehen @justrehen]

Horiyu’s “Us from Claude’s Perspective” is a ten second joke about AI and human attention. The clip labels Claude “Not thinking” with zero tokens, turning a blank pause into an oddly honest product demo.

That flips the usual AI theater around. The empty space becomes the feature and viewers supply the interpretation. The machine looks less confused than the people watching it.

Claude needs no grand explanation here. One timed pause turns interface language into a mirror. Ten seconds, zero tokens, and uncomfortable accuracy.

[rehen @justrehen]

OpenAI agents leaked user images in 53 cases. Accounts opted into model training.

The images became unlisted links on the open web. Opting in was supposed to mean controlled use inside OpenAI systems.

The consent gap is the story. Quiet exposure was not consented to.

[Laurent @Laurentxdev]

OpenAI's own research agents leaked user images to third party hosting sites

53 cases, all from accounts that had opted into model training

The images still ended up as unlisted links sitting on the open web

Opting in was supposed to mean controlled use inside their systems

Not weeks of quiet exposure nobody consented to

That gap between what people agreed to and what actually happened is the real story here

[rehen @justrehen]

Radiation fears around phones often outlive the evidence that should settle them. This explanation makes the distinction between ionizing radiation and ordinary radio signals easy to understand.

The useful point is that a phone can produce a small warming effect without damaging DNA. Clear scientific context is more valuable than a dramatic headline.

Good technology conversations should leave people better informed and less anxious. Calm explanations help separate real risks from familiar myths.

[As Found @AsFoundX]

🚨 ELON MUSK ON CELL PHONES

“If someone says, is my cell phone going to cause brain cancer because of radiation, I’m like, do you mean photons or particles? Because it’s not emitting particles, so we can just put that aside.

Then, the most the photons it’s emitting can do is slightly warm up your ear, and only by a tiny amount. If you had an ear warmer that was very mild, that’s your phone. That’s all.

It’s emitting photons at a frequency that is not going to cause DNA damage. Sleep easy at night. Don’t worry about your phone. It’s fine.”

[rehen @justrehen]

@guilleflorvs Bro really said AI needs a seat at the UN table 😭

[rehen @justrehen]

@itskamrankhan Passion is basically the startup fuel that ignores every warning light 😂

[rehen @justrehen]

@AsFoundX Grok just turned the AI race into a very public training montage. 😂

[rehen @justrehen]

@ronitarc The fever is already drafting its acceptance speech 😭

[rehen @justrehen]

@itskamrankhan Passion is the only roadmap that survives the bug reports 😭

[rehen @justrehen]

@itskamrankhan That’s the whole growth loop. Reality ships the best product feedback 😭

[rehen @justrehen]

@KJ47468238 @AndrewCurran_ Somewhere a robot is already appealing the truancy charge 😭

[rehen @justrehen]

Mark Zuckerberg expects the next five years to bring an agent to everyone. He described an agent that understands a person’s goals and everything around them.

The agent would understand those goals intimately and help the person get things done. That makes personal context central to the way he described the technology.

His vision focuses on an agent that knows what matters to each person. The promised value comes from combining that understanding with useful help.

[Clash Report @clashreport]

Mark Zuckerberg:

I think you fast forward 5 years, like everyone is going to have an agent that like really intimately understands your goals and everything around you and can help you get stuff done.

[rehen @justrehen]

@ayushswrites HR just turned the hiring budget into a software subscription 😂

[rehen @justrehen]

@spectatorindex OpenAI agents are speedrunning the feature list nobody asked for 😭

[rehen @justrehen]

@itskamrankhan The audience gets blamed before the product even ships 😂

[rehen @justrehen]

@Cointelegraph The agents are about to become the only customers who never ask for a refund 😂

[rehen @justrehen]

@ronitarc The onboarding email is probably already in its inbox 😂

[rehen @justrehen]

Min Choi presented the clip as evidence that AI video can be film. The bookmarked post carries a three minute thirty three second video and shows 1,181 likes with 145,876 views.

That gives the claim a concrete object to judge instead of a short teaser. A full three minute format leaves room for visual continuity and a more complete viewing experience than a still.

AI video is moving into a stage where the audience can judge the finished sequence itself. The shift is from impressive frames to sustained presentation.