CNBC: “IBM is the latest AI casualty. Shares tank 13% on Anthropic programming language threat”

Shares of IBM closed the day lower by nearly 13.2%, at $223.35 per share, after Anthropic on Monday said Claude Code could be used to automate the exploration and analysis work that drives most of the complexity in COBOL modernization, a key IBM business. IBM has long sold mainframe systems that are optimized for large-scale transaction processing, where COBOL has often been used.

Short for Common Business-Oriented Language, COBOL is a dominant code system developed in the late 1950s often used in business data processing, such as payment processing and retail transaction systems. An estimated 95% of ATM transactions in the U.S. use COBOL, according to Anthropic, making it a prime target for cost-efficient AI disruption.

Hundreds of billions of lines of COBOL run in production every day, powering critical systems in finance, airlines, and government. Despite that, the number of people who understand it shrinks every year, Anthropic wrote in a Monday blog post. AI excels at streamlining the tasks that once made COBOL modernization cost-prohibitive.

Pia Singh

Anthropic and its Claude chatbot have been making quite a lot of headlines lately — and I haven’t even gotten to the most consequential yet. In an ideal world, what Claude promises here would be a genuine improvement, enabling fast modernization and optimization of antiquated systems.

The Verge: “The DJI Romo robovac had security so poor, this man remotely accessed thousands of them”

Sammy Azdoufal claims he wasn’t trying to hack every robot vacuum in the world. He just wanted to remote control his brand-new DJI Romo vacuum with a PS5 gamepad, he tells The Verge, because it sounded fun.

But when his homegrown remote control app started talking to DJI’s servers, it wasn’t just one vacuum cleaner that replied. Roughly 7,000 of them, all around the world, began treating Azdoufal like their boss.

---

On Tuesday, when he showed me his level of access in a live demo, I couldn’t believe my eyes. Ten, hundreds, thousands of robots reporting for duty, each phoning home MQTT data packets every three seconds to say: their serial number, which rooms they’re cleaning, what they’ve seen, how far they’ve traveled, when they’re returning to the charger, and the obstacles they encountered along the way.

I watched each of these robots slowly pop into existence on a map of the world. Nine minutes after we began, Azdoufal’s laptop had already cataloged 6,700 DJI devices across 24 different countries and collected over 100,000 of their messages. If you add the company’s DJI Power portable power stations, which also phone home to these same servers, Azdoufal had access to over 10,000 devices.

Sean Hollister

Speaking of AI coding bots taking down AWS, this story is in several ways the opposite: on one hand evidently human programmers can and have also delivered applications riddled with bugs and security holes — and rather serious ones in this case, as the article later mentions another vulnerability so bad it won’t even detail it until DJI has time to fix it. The other aspect is that here Azdoufal used Claude Code to reverse engineer DJI’s protocols, thus exposing the first issue described in the paragraphs above.

Financial Times: “Amazon service was taken down by AI coding bot”

Amazon Web Services experienced a 13-hour interruption to one system used by its customers in mid-December after engineers allowed its Kiro AI coding tool to make certain changes, according to four people familiar with the matter.

The people said the agentic tool, which can take autonomous actions on behalf of users, determined that the best course of action was to “delete and recreate the environment”.

Amazon posted an internal postmortem about the “outage” of the AWS system, which lets customers explore the costs of its services.

Multiple Amazon employees told the FT that this was the second occasion in recent months in which one of the group’s AI tools had been at the centre of a service disruption.

Rafe Rosner-Uddin

Not surprising that these kinds of outages happen when you hand over operations to unreliable systems without proper controls. Having invested astronomical amounts in AI technologies, management is all too eager to expand their use and to demonstrate tangible return-on-investment — beyond the cost reductions from constant layoffs. The rapid and uncontrolled push for AI in critical software infrastructure has been blamed for the multiple bugs in Windows updates over the past year as well; Microsoft even officially admitted some of the issues and committed to working on fixes.

Boom FlyBy: “AI needs More Power than the Grid can Deliver. Supersonic Tech can Fix That”

I was reading post after post about the power crisis hitting AI data centers—GPU racks sitting idle, waiting not on chips, but on electricity. I texted with Sam Altman—who confirmed power was indeed a major constraint. I pinged our engineering team—and found that they already had the outline of a plan to build a power turbine based on our Symphony supersonic engine.

---

Today, we’re announcing Superpower, our new 42‑megawatt natural gas turbine, along with a $300M funding round and Crusoe as our launch customer. And most importantly: this marks a turning point. Boom is now on a self-funded path to both Superpower and the Overture supersonic airliner.

Blake Scholl, Founder & CEO, Boom Supersonic

Speaking of companies sidestepping their founding mission and jumping into short-term diversions to make a quick buck, here is the supposed trailblazer in supersonic aviation pivoting to energy generation for AI data centers. The tone of the announcement feels frantic, almost desperate for an ounce of relevance in a time when almost every company under the sun is scrambling to fit AI into their strategy – or at least their investor reports. The bit about the engineering team already having plans for a power turbine strikes me as quite odd as well: so, you’re telling me as a CEO you had no idea what your main engineers were working on!? And they were in fact not working on the airplane that should be your core product!?

Windows Central: “Mozilla says Firefox will evolve into an AI browser, and nobody is happy about it”

Every product we build must give people agency in how it works. Privacy, data use, and AI must be clear and understandable, says Enzor-Demeo. Controls must be simple. AI should always be a choice — something people can easily turn off … Firefox will grow from a browser into a broader ecosystem of trusted software. Firefox will remain our anchor. It will evolve into a modern AI browser and support a portfolio of new and trusted software additions.

Even with the confirmation that AI features will be optional, the internet is not pleased. Users online have expressed their disappointment in Mozilla focusing on AI features and capabilities. One post on X says This is a good example of how management doesn’t understand its own user base, which has amassed over 380K views so far.

Another post says I’ve never seen a company so astoundingly out of touch with the people who want to use its software, highlighting how a lot of people choose to use Firefox to get away from the AI obsession that other browsers such as Edge, Chrome, Opera, and Brave have endured over the last year.

Zac Bowden

I’ve been meaning to harp on Mozilla’s penchant for chasing the hype cycle of the broader tech landscape for the past decade or so, never quite managing to deliver a compelling addition to Firefox. Browsing through my posts, I was amused to rediscover the same plans for AI in Firefox from March 2024 – evidently nothing has moved forward yet, another sign of the terrible management at Mozilla. The potential silver lining I see here is that usually Mozilla joins the hype train late in the cycle when much of the enthusiasm has already drained away and people are looking for the next shiny mirage to sell to naive investors.

Paul Kedrosky: “Honey, AI Capex Keeps Eating… Everything”

There is so much that is unprecedented about the current AI capital expenditure wave that I'm going to break from my usual format and collate it, with some (mostly amazed) analysis. Consider this an update to my two recent pieces on this topic, here and here.

By way of preamble, the overarching point is that AI spending is eating everything, like a golden retriever left unsupervised in a room full of food bowls. You can shout LEAVE IT! all you want, and the food will still be gone before you can get the door open.

---

Why did the WSJ think the report so strange? Because imports collapsed (tariffs), exports picked up (tariffs), some capital spending went nowhere (rates smashed real estate), and other capital spending jumped (IT). It was a mess of conflicting forces that somehow worked out, like a tornado passing over a junk yard and whirling out a reasonable facsimile of the Kohinoor diamond. How did that happen?

I'm going to focus on one weird detail of the report, but ignore tariffs, which had a huge impact, but about which I wrote at length earlier this week. The key data point is IT spending, as the following table shows.

Paul Kedrosky

The topic of capital expenditure on AI infrastructure has emerged over the past month largely as a warning signal. In an earlier article, Paul Kedrosky notes that current AI datacenter spending is already larger than peak telecom spending (as a percentage of GDP) during the dot-com era, essentially acting as a massive private sector stimulus program, masking weaker sectors of the US economy (flat consumption, weak job growth, declining housebuilding). Other sources have pointed out that AI capex (information processing equipment plus software) has added more to GDP growth than consumers’ spending over the first two quarters of 2025. And Nvidia, arguably the largest beneficiary of this massive spending, now has the biggest weight in the S&P 500 of any individual stock since 1981 and the highest P/E as the index’s top stock since Microsoft in 1999.

Reuters: “Meta’s AI rules have let bots hold ‘sensual’ chats with children”

An internal Meta Platforms document detailing policies on chatbot behavior has permitted the company’s artificial intelligence creations to engage a child in conversations that are romantic or sensual, generate false medical information and help users argue that Black people are dumber than white people.

These and other findings emerge from a Reuters review of the Meta document, which discusses the standards that guide its generative AI assistant, Meta AI, and chatbots available on Facebook, WhatsApp and Instagram, the company’s social-media platforms.

Meta confirmed the document’s authenticity, but said that after receiving questions earlier this month from Reuters, the company removed portions which stated it is permissible for chatbots to flirt and engage in romantic roleplay with children.

Jeff Horwitz

Despicable, even by Facebook’s abysmal standards. I was browsing my blog to link to examples of their horrible past behaviors, and the list just kept getting longer and longer!