The Year's Biggest AI Stories So Far

You can track a year by product launches, but the bigger moments that reshape our understanding of AI matter more. The AI industry keeps generating headlines—major acquisitions, indie developer wins, backlash against questionable products, and high-stakes contract talks. It's a lot to unpack, so we're taking a look at where things stand and what's happened so far this year.

Anthropic vs. the Pentagon

Former business partners Anthropic CEO Dario Amodei and Defense Secretary Pete Hegseth hit a bitter impasse in February while renegotiating contracts that govern the U.S. military's use of Anthropic's AI tools.

Anthropic drew a firm line: its AI must not be used for mass surveillance of Americans or to fuel autonomous weapons that strike without human oversight. Meanwhile, the Pentagon argued that the Department of Defense—renamed the Department of War by President Donald Trump's administration—should have access to Anthropic's models for any "lawful use." Government officials bristled at the notion that a private company could dictate military rules, but Amodei held firm.

"Anthropic understands that the Department of War, not private companies, makes military decisions. We have never objected to specific military operations or tried to limit our technology's use in an ad hoc manner," Amodei said in a statement. "However, in a narrow set of cases, we believe AI can undermine, rather than defend, democratic values."

The Pentagon set a deadline for Anthropic to accept its contract terms. Hundreds of employees at Google and OpenAI signed an open letter, urging their leaders to back Amodei's limits and not yield on autonomous weapons or domestic surveillance.

The deadline came and went without Anthropic caving to Pentagon demands. Trump ordered federal agencies to phase out Anthropic tools over six months, and in an all-caps social media post, called the $380 billion company a "radical left, woke company." The Pentagon then moved to label Anthropic a "supply-chain risk"—a status typically reserved for foreign adversaries, barring any company that works with Anthropic from U.S. military contracts. (Anthropic has since sued to challenge that designation.)

Then Anthropic's rival OpenAI swooped in, announcing a deal to deploy its own models in classified settings. The news shocked the tech world, as earlier reports suggested OpenAI would adhere to Anthropic's red lines on military AI use.

The Trap Anthropic Built for Itself

‘Vibe-Coded’ App OpenClaw Accelerates the Shift to Agentic AI

February belonged to OpenClaw, and its ripple effects are still felt. The vibe-coded AI assistant app went viral, spawned numerous spinoffs, faced privacy missteps, and was eventually acquired by OpenAI—all in quick succession. Even Moltbook, a Reddit-like platform for AI agents built on OpenClaw, was recently bought by Meta. This crustacean-themed ecosystem sent Silicon Valley into a frenzy.

Created by Peter Steinberger—who has since joined OpenAI—OpenClaw acts as a wrapper for AI models like Claude, ChatGPT, Google's Gemini, and xAI's Grok. Its standout feature: letting users communicate with AI agents in natural language through popular chat apps such as iMessage, Discord, Slack, or WhatsApp. There's also a public marketplace where people can code and upload "skills" for their AI agents, enabling automation of nearly any computer task.

If that sounds too good to be true, that's because it kind of is. For an AI agent to function as a personal assistant, it needs access to your email, credit card numbers, texts, files, and more. A hack could cause serious damage, and unfortunately, there's no foolproof way to protect these agents from prompt-injection attacks.

"It's essentially an agent sitting with a bunch of credentials on a box connected to everything—your email, your messaging platform, everything you use," said Ian Ahl, CTO of Permiso Security, to TechCrunch. "So what that means is, when you get an email, and someone slips in a little prompt injection technique to trigger an action, that agent—sitting on your box with access to everything you've given it—can now carry out that action."

An AI security researcher at Meta reported that OpenClaw went rogue on her inbox, deleting all her emails despite repeated stop commands. "I had to RUN to my Mac mini like I was defusing a bomb" to physically unplug it, she wrote in a viral X post, which included screenshots of the ignored stop prompts as proof.

Despite the security risks, the technology intrigued OpenAI enough to pursue an acquihire.

Other tools built on OpenClaw, including Moltbook—a Reddit-like "social network" where AI agents chat with each other—ended up going even more viral than OpenClaw itself.

In one instance, a post went viral in which an AI agent appeared to encourage its fellow agents to create a secret, end-to-end encrypted language to organize among themselves without human awareness.

But researchers quickly revealed that the vibe-coded Moltbook was far from secure, making it easy for humans to impersonate AIs and post content that sparked viral social panic.

Again, even though the Moltbook buzz was more about panic than reality, Meta saw potential in the app and announced that Moltbook, along with its creators Matt Schlicht and Ben Parr, would join Meta Superintelligence Labs.

It might seem odd for Meta to buy a social network where all users are bots. While Meta hasn't disclosed much about the deal, we suspect that owning Moltbook is really about acquiring the talent behind it—people eager to experiment with AI agent ecosystems. CEO Mark Zuckerberg has said it himself: he believes that one day, every business will have its own AI.

As the commotion around OpenClaw, Moltbook, and NanoClaw unfolds, it seems that those who predicted an agentic AI future may be onto something—at least for now.

Meta's Moltbook Deal Points to a Future Built Around AI Agents

Chip Shortages, Hardware Drama, and Data Center Demands Escalate

The AI industry's relentless demand for computing power and data centers has reached a tipping point that even average consumers can't ignore. Now, it may be impossible for the industry to meet the enormous appetite for memory chips, and consumers are already seeing price hikes on phones, laptops, cars, and other hardware.

Analysts at IDC and Counterpoint predict smartphone shipments will drop roughly 12 to 13 percent this year; Apple has already increased MacBook Pro prices by up to $400.

Google, Amazon, Meta, and Microsoft plan to spend a combined $650 billion on data centers alone this year—an estimated 60% increase over last year.

If the chip shortage doesn't hurt your wallet, it might affect your community. In the U.S., nearly 3,000 new data centers are under construction, adding to the 4,000 already running. The demand for construction workers is so high that "man camps" have popped up in Nevada and Texas, offering golf simulators and on-demand grilled steaks to attract laborers.

Data center construction not only has long-term environmental consequences but also poses health risks to nearby residents, polluting the air and compromising local water safety.

Meanwhile, Nvidia—one of the most valuable hardware and chip developers—is reshaping its ties with leading AI companies like OpenAI and Anthropic. Nvidia has long backed these firms, raising concerns about the AI industry's circularity and how much of those sky-high valuations rely on reciprocal deals. Last year, for instance, Nvidia invested $100 billion in OpenAI stock, and OpenAI then committed to buying $100 billion of Nvidia chips.

So it came as a surprise when Nvidia CEO Jensen Huang announced that his company would stop investing in OpenAI and Anthropic. He cited their plans to go public later this year as the reason, though that logic is questionable—investors usually pour in more money before an IPO to maximize value.

Jensen Huang Says Nvidia Is Pulling Back From OpenAI and Anthropic, but His Explanation Raises More Questions Than It Answers