Tech
'We're so used to relying on technology.' Hospitals, airlines, agency operations around New England hampered by software outage.
Blue screen of death on display (David Ryan/Getty Images)

When in doubt, blame Europe: Microsoft pins fault for CrowdStrike fiasco on the EU

J'accuse!

If you were unfortunate enough to spend last weekend in an airport, your travel plans were probably wrecked by a widespread “blue screen of death” mishap that shut down 8.5 million Windows-powered devices, causing US airlines to cancel more than 5,000 flights on Friday and Saturday. The cause of the computer outage was a faulty software update pushed by cybersecurity provider Crowdstrike to Microsoft devices, and Microsoft blamed the system vulnerability that caused this software issue on a 2009 agreement with the EU. From The Telegraph:

Microsoft has blamed EU rules for enabling a faulty security update to cause the world’s biggest IT outage. The software giant said a 2009 agreement with the European Commission meant it was unable to make security changes that would have blocked the CrowdStrike update that triggered widespread travel and healthcare chaos on Friday.

CrowdStrike’s Falcon system, designed to prevent cyber attacks, has privileged access to a key part of a computer known as the kernel. This meant that a faulty update last week resulted in millions of Windows computers and servers being unable to load at all, leading to flight cancellations, contactless payments not working and GP surgeries being unable to make appointments.

Microsoft, which offers its own alternative to CrowdStrike known as Windows Defender, agreed in 2009 to allow multiple security providers to install software at the kernel level amid a European competition investigation.

In contrast, Apple blocked access to the kernel on its Mac computers in 2020, which it said would improve security and reliability. A Microsoft spokesman told the Wall Street Journal that it was unable to make a similar change because of the EU agreement.

For context, the “kernel” is a computer program at the core of its operating system, and buggy software updates that interact with an operating system’s kernel can, as we saw with Crowdstrike, wreak havoc on devices using that OS. Apple runs a closed operating system, locking third-party software providers out of its kernel, which safeguards its devices from incidents like this.

I wrote last week about the EU’s obsession with obscene fines for US big tech companies, so it’s fitting that Microsoft is now blaming European regulators for its 8.5 million device failure. Ben Thompson provided excellent background to Microsoft’s explanation:

Two of the companies seizing this opportunity in the 2000s were Symantec and McAfee; both reacted with outrage in 2005 and 2006 when Microsoft, in the run-up to the release of Windows Vista, introduced PatchGuard. PatchGuard was aptly named: it guarded the kernel from being patched by 3rd-parties, with the goal of increasing security…

Symantec, meanwhile, went straight to E.U. regulators, making the case that Microsoft, already in trouble over its inclusion of Internet Explorer in the 90s, and Windows Media Player in the early 2000s, was unfairly limiting competition for security offerings. The E.U. agreed and Microsoft soon backed down.

Basically, Microsoft wanted to lock third-party security software providers out of its kernel, two of said software providers cried “anti-competitive!” EU regulators agreed, and Microsoft dropped its efforts to block kernel access. Eighteen years later, a security software provider with kernel access pushed an update that shutdown millions of computers, which wouldn’t have happened if that software provider didn’t have kernel access. The EU, for what it's worth, denied responsibility for the computer failure.

Considering that 72% of global desktop computers run on Microsoft operating systems, it will be interesting to see if the tech giant can leverage the Crowdstrike bug to reverse the EU’s open-OS stance on Windows.

More Tech

See all Tech
tech

Report: Some of Meta’s new AI models will eventually be open-source

Axios reports that Meta is close to releasing its first new AI models after setting up its “superintelligence” team led by former Scale.AI CEO Alexandr Wang, and some of the models will eventually be released with an open-source license.

Per the report, Meta sees an opportunity to focus on consumers, rather than the lucrative enterprise market that both OpenAI and Anthropic have been focusing on.

Meta had previously embraced open-source AI with its Llama models, with CEO Mark Zuckerberg writing a manifesto declaring open-source AI as “the path forward.” Axios says that Meta will be pursuing more of a hybrid strategy of proprietary and open-source models going forward.

The New York Times previously reported that Meta was delaying the launch of its new AI model because of performance issues.

Meta had previously embraced open-source AI with its Llama models, with CEO Mark Zuckerberg writing a manifesto declaring open-source AI as “the path forward.” Axios says that Meta will be pursuing more of a hybrid strategy of proprietary and open-source models going forward.

The New York Times previously reported that Meta was delaying the launch of its new AI model because of performance issues.

1328213286	CSA-Printstock

OpenAI’s plan for an AGI world: AI for all and a 4-day workweek

The company’s policy paper calls for a new social contract that includes AI at the center of everything, which could lower costs and create cures for diseases, but also warned it may upend the public safety net.

🏠 $2.15M

The median price for a house in San Francisco is now $2.15 million, jumping 18% from last year. The AI startup boom is pushing what was already one of the most expensive housing markets to dizzying new heights. The median price for condos in the city jumped 27% to reach $1.36 million, according to data from Compass, reported by Bloomberg.

tech

Report: OpenAI on track to burn $85 billion in 2028, expects profitability by 2030

Anthropic and OpenAI are racing to go public this year, and all eyes are on how long they can sustain burning billions in cash before they achieve something that looks like a viable business.

Investors have seen both companies’ projections, and there’s no sign of slowing down, according to a report from The Wall Street Journal.

OpenAI expects to burn tens of billions per year for the rest of the decade, peaking at $85 billion in 2028, before achieving profitability in 2030, per the report.

Anthropic will also continue to burn cash for years — far less than OpenAI — but it projects that 2026 will be its biggest year of losses. It targets 2029 for profitability, fueled by exploding enterprise revenue.

OpenAI expects to burn tens of billions per year for the rest of the decade, peaking at $85 billion in 2028, before achieving profitability in 2030, per the report.

Anthropic will also continue to burn cash for years — far less than OpenAI — but it projects that 2026 will be its biggest year of losses. It targets 2029 for profitability, fueled by exploding enterprise revenue.

Form Energy iron-air battery system leaving Form Factory 1

Big batteries are the newest answer to Big Tech’s big energy needs

America’s booming energy demand is creating a powerful case for large-scale energy storage.

Patrick Sisson4/2/26

Latest Stories

Sherwood Media, LLC produces fresh and unique perspectives on topical financial news and is a fully owned subsidiary of Robinhood Markets, Inc., and any views expressed here do not necessarily reflect the views of any other Robinhood affiliate, including Robinhood Markets, Inc., Robinhood Financial LLC, Robinhood Securities, LLC, Robinhood Crypto, LLC, Robinhood Derivatives, LLC, or Robinhood Money, LLC. Futures and event contracts are offered through Robinhood Derivatives, LLC.