One of the fathers of cyberpunk said that “tech bros” are constantly thanking him for the inspiration.

For context, Gibson wrote about the abuse of technology and corporations controlling the world instead of governments.

Now Silicon Valley CEOs see his books as a utopia.

It’s about to get harder for IT workers to move to the US: Microsoft, Adobe and several other Big Tech companies have been barred from hiring foreign nationals through the green card program.

The US Department of Labor will halt the processing of all existing and new applications. And employees of these companies on H-1B visas will now simply be denied US work-based permanent residency.

Next, the White House wants to go after foreign nationals at Harvard, Yale and Stanford.

Everything about Claude Code in one place: a detailed course on working with the AI agent is now available.

It takes you from prompting principles and working with Git to building wrappers, launching subagents, and creating your own skills and plugins. The advanced track teaches you how to orchestrate multiple agents simultaneously.

There are practical assignments with feedback throughout. The course is regularly updated for the latest versions of Claude Code.

Study for free via this link.

Bill Gates has warned of the risk of mass AI cyberattacks as early as 2027. According to him, “we’re in for a lot of unpleasant surprises.”

A few more quotes from the Microsoft founder’s recent CNN interview:

There will be grim consequences if we don’t maintain control

Over the next year, we’ll go beyond
self-governance

Obviously, we’re not doing enough

Well then.

AI developers are taking up smoking—all because they believe AI will cure cancer, says Palantir CTO Shyam Sankar.

Many AI researchers have started smoking to show their colleagues just how much they believe in AI. So they’re putting everything on the line

Sankar didn’t name any specific people or figures. Maybe it’s still too early to believe in AI that much.

Epic Games are making a game with “not a single line of hand-written code” — instead of developers, neural networks do all the writing. That’s how the studio is working on a shooter featuring Marvel and Star Wars heroes.

The development team was told to create all the graphics and code entirely through Claude Code. Humans were left with nothing but managerial roles.

The release was planned for March 2027, but the project has now been postponed indefinitely. The new requirements caught employees off guard.

May Claude be with them.

The world failed to agree on how to save Earth: scientists simulated negotiations between countries.

They discussed spraying particles into the atmosphere to reflect sunlight. Here’s what came of it:

– The US and China immediately refused to cooperate on anything that harmed their national interests;
– The Americans then “exposed” China’s secret experiments and sabotaged the deal, even though China wasn’t conducting any experiments according to the scenario;
– A fake environmental NGO funded by oil companies set up the game’s scientific adviser — although hardly anyone was listening to her anyway;
– By the end of the experiment, the real scientist had been reduced to tears, and the treaty between the countries still hadn’t been signed.

Was this really a simulation?

No more need for an Adobe subscription: a Redditor vibe-coded a Lightroom alternative. We tested it — it has the potential to become a viable replacement.

It includes all the basic photo-editing tools, preset and RAW support, object removal, and photo library management. You can connect it to Claude Code and do everything with prompts.

Try it for free via the link. The coolest part is that the developer also released a Photoshop clone. After Effects and Illustrator are next.

Engineers have come up with invisible tattoos — they only appear under ultraviolet light.

In sunlight, the tattoo will show up in an “intermediate” form. Ordinary bright lamps hide it instantly.

The authors suggest “turning off” the tattoo before visiting your grandmother or attending important meetings.

Anthropic and OpenAI are preparing for a major “AI catastrophe”: AI developers are working through scenarios in which society and governments turn against them.

Axios insiders say a global cyberattack is the most likely scenario — it could knock out banks, the internet, and even public utilities. Such a crisis is expected within the next 6–12 months.

After that, attempts could be made to ban AI or freeze its development. That’s why the companies are already working with politicians to influence future laws.