That Little Logo Does a Lot of Heavy Lifting
You've seen the UL mark thousands of times — a simple circle with two letters printed on the back of your blender, your phone charger, your laptop adapter. It's one of those symbols so ubiquitous that most people never stop to think about what it actually means, or the century-old organization behind it.
Underwriters Laboratories, now operating as UL Solutions, has been in the business of safety testing since the early days of electrification. When electricity first started entering homes in the late 19th century, insurance companies needed a reliable way to know which products were safe enough to insure. UL was the answer — a neutral third-party lab that could put products through rigorous fire and electrical testing and give insurers (and eventually consumers) a trusted signal.
That mandate hasn't shrunk. If anything, it's exploded.
A Business Hidden in Plain Sight
In a recent episode of the Decoder podcast, host Nilay Patel sat down with UL Solutions CEO Jennifer Scanlon to pull back the curtain on a company that most people interact with daily without knowing it. The conversation touches on something genuinely interesting: how do you run a safety certification business in an era where the products being certified are getting more complex, cheaper, and more globally distributed by the day?
Scanlon points to a real tension at the heart of UL's market: the explosion of inexpensive electronics on platforms like Amazon has created a consumer environment where price often wins over certification. Not every seller on a third-party marketplace bothers to get UL-listed, and not every buyer stops to check. The mark that once served as a near-mandatory stamp of legitimacy is competing in a world where speed-to-market and low prices can override safety considerations.
That's a structural challenge UL is grappling with — and it only gets harder from here.
Can You Safety-Test an AI?
The most forward-looking part of UL Solutions' work right now is an attempt to bring its century-old safety-testing philosophy to artificial intelligence. The company recently rolled out UL 3115, described as "a structured framework to evaluate AI-based products before and during deployment."
It's an ambitious move — and a philosophically tricky one. Traditional safety testing works because a toaster behaves the same way every time under controlled conditions. AI systems, by contrast, are probabilistic, context-dependent, and constantly evolving. What does it even mean to certify something as "safe" when its outputs can shift based on data, user interaction, or model updates?
For UL 3115 to gain traction, it needs buy-in from manufacturers, regulators, and the broader tech industry — none of which are known for moving in lockstep. Scanlon acknowledges the challenge: building a credible AI safety standard requires not just a methodology, but trust from the companies being evaluated and the governments that would ultimately rely on those certifications.
Why This Matters Beyond the Label
UL Solutions' story is a useful lens for thinking about how safety infrastructure actually gets built — and how it can be left behind by the pace of technological change. The company spent decades establishing legitimacy in the physical world. Now it's betting that same institutional credibility can transfer to the digital and algorithmic one.
Whether that bet pays off will depend on whether regulators, manufacturers, and consumers decide that AI safety certifications are worth something — or treat them the way many already treat the UL mark on a $4 phone charger from an unknown seller.
Source: The Verge / Decoder podcast, featuring UL Solutions CEO Jennifer Scanlon.
