🤖 Anthropic Is Making Its Own Drugs — And That's Just the Start
Welcome to Daily Inference, your daily dose of the most important AI stories shaping our world. I'm glad you're here, because today's lineup is packed — we're talking AI going pharmaceutical, OpenAI cozying up to Washington, a serious warning for every parent online, Anthropic's silicon ambitions, and the growing question of whether AI's direction has already been decided without any of us having a say. Let's get into it.
Before we dive in, a quick word from our sponsor, 60sec.site — an AI-powered tool that lets you build a stunning website in, you guessed it, about sixty seconds. Whether you're launching a business, a portfolio, or a side project, 60sec.site makes it ridiculously easy. Check them out at 60sec.site.
Okay, first up — Anthropic is making a bold pivot that caught a lot of people off guard. The company, best known for its Claude AI models and its dominance in coding tools, has announced something called Claude Science — essentially an all-in-one workbench for researchers that unifies fragmented scientific tools and datasets into a single environment. But here's where it gets really interesting: Anthropic didn't stop there. The company is now saying it wants to develop its own drugs. That's right — not just power pharma research for other companies, but actually get into the business of creating healthcare interventions itself. With a long list of biotech and pharma customers already using Claude, this feels less like a moonshot and more like a deliberate strategic move. And this story connects neatly to something else happening in AI hardware — Anthropic is reportedly in discussions with Samsung to develop a custom AI chip. Just a week after OpenAI announced its own chip partnership with Broadcom, Anthropic seems determined to control more of its own technological destiny, from silicon all the way up to drug pipelines. The AI arms race is no longer just about model performance — it's about vertical integration at every layer.
Speaking of OpenAI and strategic maneuvering — Sam Altman has reportedly proposed handing five percent of OpenAI's equity to a US sovereign wealth fund, essentially giving the American public a financial stake in the AI boom. Based on OpenAI's most recent valuation of around 852 billion dollars, that slice would be worth a staggering amount. According to reporting from the Financial Times, Altman first floated this idea to the Trump administration early last year, framing it as the best way to share AI's upside with the public while simultaneously easing political tensions. It's a fascinating move — part public relations strategy, part genuine policy proposal. And it comes at a moment when the question of who benefits from AI's massive wealth creation is becoming impossible to ignore. Over in South Korea, chipmakers like Samsung and SK Hynix are generating three thousand percent bonuses for some workers, yet analysts are raising serious questions about how unevenly that wealth is being distributed. Meanwhile, in Australia, tech giants are reportedly lobbying to weaken copyright laws in exchange for fifty billion dollars in data center investment — a deal that musicians and creatives are calling the ultimate dirty deal. The pattern is clear: AI is generating enormous value, but the fights over who captures that value are just getting started.
Now let's talk about something that every parent needs to hear. The UK's National Crime Agency and the Internet Watch Foundation have issued landmark guidance warning parents not to publicly post photos of their children online. The reason is alarming — so-called nudification apps, powered by advances in AI imaging, are being used by predators to take ordinary photos of children and transform them into explicit material. What makes this especially disturbing is that children can become victims without ever having direct contact with criminals. The imagery tools have become so sophisticated and widely available that a simple social media post can become dangerous. The guidance suggests parents either make their accounts private or share children's photos exclusively through close-friends groups. This story is a stark reminder that AI capabilities don't exist in a vacuum — they get weaponized, and the most vulnerable pay the price when safeguards lag behind the technology.
On a different but equally important note — can you tell whether a piece of writing was created by a human or an AI? That question is moving from philosophical thought experiment to urgent practical reality. Linguists and novelists including Jennifer Egan and Jeanette Winterson are now wrestling with what truly distinguishes human writing from machine-generated text. The answer, it turns out, is more nuanced than most people assume. AI language tends to be fluent, even eloquent — but it often lacks the idiosyncratic friction, the emotional specificity, the sense of a life behind the words that defines great human writing. And yet, allegations of undisclosed AI use are now surfacing across the literary and media worlds. This connects to a broader anxiety: as Mark Zuckerberg reportedly told Meta staff in an internal meeting that AI agents haven't progressed as quickly as he'd hoped, we're reminded that AI still has very real limitations. But those limitations are shrinking fast — and society's ability to detect and adapt is struggling to keep pace.
Finally, let's zoom out to the big philosophical picture, because it matters. A fascinating public debate is unfolding around whether AI's ethical direction has already been quietly predetermined — not by philosophers or policymakers, but by the enormous financial incentives driving the industry. Hundreds of billions of dollars have been committed to AI development because it promises commercial returns and geopolitical advantage. Those pressures, as one Guardian reader put it so sharply, are quietly determining the future before society has consciously decided where it wants to go. Meanwhile, at Google DeepMind, unionization talks between employees and executives have gotten off to a rocky start, with workers voicing frustration at what they see as executives unwilling to genuinely engage with labor concerns. It's a telling tension — the people building some of the most powerful AI systems in the world are themselves asking for more say in how that power is exercised. And that feels like exactly the kind of question we should all be paying more attention to.
That's your Daily Inference for today. A lot happening, a lot to think about — and honestly, that's pretty much every day in AI right now. If you want to stay on top of it all, head over to dailyinference.com and sign up for our daily AI newsletter. We break it all down, five days a week, straight to your inbox. And don't forget to check out today's sponsor, 60sec.site, for AI-powered website building that actually works. Until next time — stay curious, stay informed, and we'll see you tomorrow.
