Anthropic wants to develop its own drugs
Anthropic launched “Claude Science” — a workbench for researchers that unifies tools and datasets and generates scientific figures — and then casually announced it wants to develop its own drugs. That’s a notable pivot for a company whose whole brand is “we’re the responsible AI lab,” but sure, let’s add pharmaceutical development to the list. The ambition is real, and AI-assisted drug discovery is genuinely promising, but I’ll be watching closely to see whether this is a research moonshot or a valuation story.
Amazon will stop accepting new customers for Mechanical Turk
Mechanical Turk — Amazon’s crowdsourced human-labor marketplace, which spent two decades quietly powering AI training datasets — is effectively being wound down. This is a bigger deal than it sounds: MTurk was the unsexy backbone of the AI revolution, the place where humans labeled millions of images, transcribed audio, and did the unglamorous work that made “intelligent” systems possible. Its sunset is a statement about where AI training is heading, and not everyone’s going to like the answer.
Alibaba reportedly bans employees from using Claude Code
Alibaba has classified Claude Code as “high-risk software” and reportedly banned employees from using it — which, given that Claude Code is an Anthropic product and Anthropic is backed by Google and Amazon, has layers of corporate geopolitics baked in. Data security concerns are the stated reason, and they’re not unreasonable for an AI tool that has deep access to your codebase. This story pairs nicely with VentureBeat’s look at open-source alternatives eating Claude Code’s lunch on price — the pressure on Anthropic’s developer tools is coming from multiple directions at once.
Claude Code costs up to $200 a month. Goose does the same thing for free.
Block’s open-source coding agent Goose is making a credible case that you don’t need to pay Anthropic $200 a month for an agentic coding tool. The pricing rebellion among developers is real and accelerating — when the people who build software start doing the math on their AI subscriptions, the economics get uncomfortable fast. This is the moment in every platform cycle where the open-source ecosystem catches up and the incumbents have to figure out what they’re actually selling.
Google just redesigned the search box for the first time in 25 years
The blinking cursor in a white rectangle is going away — Google is overhauling the search interface that has been essentially unchanged since 1998, reorienting it around AI queries and multimodal input. This is one of those changes that sounds incremental until you realize the search box is probably the most-used interface in human history. If it works, this is a genuinely historic UX shift; if it doesn’t, a billion people are going to be annoyed.
Infuriating Google commercial imagines the founding fathers embracing AI
“Group project, but make it 1776.” Google’s new Workspace commercial imagines Thomas Jefferson pinging Ben Franklin over Gemini suggestions for the Declaration of Independence, and it is exactly as cringeworthy as that sentence implies. I understand why marketing teams do this — aspirational, historical gravitas, democratizing genius — but there is something uniquely galling about implying that the document that launched a republic would have benefited from an AI sidebar. The founding fathers had one job, Google. Let them have it.
The fanfiction community is at war with AI — and itself
The AO3 fanfiction community is deploying AI detectors to root out AI-generated stories — which is a delicious irony, since AI detectors are notoriously unreliable and are probably flagging human writers as bots right now. This is a microcosm of a much larger cultural fight over authenticity, labor, and what “creative work” even means in 2026. What’s notable here is that the strongest anti-AI sentiment isn’t coming from publishing houses or studios — it’s coming from a community of passionate amateurs who just want to be the weirdos writing their own stories.
Some of the nation’s rich are letting AI teach their kids
Wealthy families are enrolling their children in AI-forward schools and tutoring programs like Forge Prep and Alpha, betting that personalized AI instruction will give their kids an edge. The equity angle writes itself — the same parents who can afford to opt their children out of whatever they don’t like are now opting them into the AI experiment with the confidence that money provides. Whether this is visionary or just expensive snake oil dressed in silicon will depend entirely on whether these tools actually work, and we won’t know that for a decade.
Bottom Line
The AI industry is simultaneously trying to teach your kids, develop your drugs, redesign the interface you’ve used since dial-up, and kill off the human labor market it was built on — all before lunch on a Monday.