Feedback is New Scientist’s a popular side view of the latest science and technology news. Items you think readers might find interesting can be submitted to Feedback by email at feedback@newscientist.com
It’s on gas
The feedback is bold, so here’s a prediction: the research we’re about to describe will win an Ig Nobel within the next decade. The whole project is tailor-made for Igs. It is an attempt to objectively measure human flatulence using biosensors, or “smart underwear”.
We learned this from a press release from the University of Maryland, brought to our attention by physics reporter Karmela Padavic-Callaghan with the line: “Sure, Feedback can do something about it.”
A key problem is that we don’t know the normal range for gas, unlike other key biomarkers like blood glucose. Most studies relied on self-report, which doesn’t really work because people often don’t remember all of their farts and misjudge how big each one was. Plus, there’s the “impossibility of detecting gas in your sleep”: anyone who’s shared a bed with someone else knows that everyone farts in their sleep.
Hence the Smart Underwear developed by Brantley Hall and colleagues. The press release calls it “a small wearable device that discreetly clips onto any undergarment and uses electrochemical sensors to continuously monitor intestinal gas production.” I wonder what constitutes “petty” in this context, Feedback ticked scientific workand it turns out that the sensor has dimensions of 26 x 29 x 9 millimeters – which, we admit, is quite small, but the participants of the experiment would like to avoid skinny jeans.
Based on the first round of studies, “healthy adults caused flatulence an average of 32 times per day,” about twice as often as previously thought. But people vary widely: daily totals ranged between four and 59 farts.
As Smart Underwear expands, the data it collects will be fed into a larger project, the Human Flatus Atlas. This has a website (flatus.info) where you can sign up to watch farts. Participants are tempted by the prospect of finding out if they’re a hydrogen hyperproducer, a Zen hood who barely farts even on a diet of baked beans or something in between.
The feedback is curious how resistant the sensors are against significant farts. We recently learned of a gentleman who visited a French hospital after being introduced unexploded shell from the First World War to its bottom, forcing personnel to operate with the help of a bomb disposal unit. We guess anything coming from this quarter might have been too much for smart underwear.
Meanwhile, the lead researchers established VentosCity use tech. Its website is minimal, just an animation of some gas, a tagline (“Measure. Master. Thrive.”) and a promise: “The future of gut health is coming.” Feedback suspects a monthly subscription app coming soon.
Ghost in the Machine
As AI companies introduce their technologies into every aspect of our lives, we need help understanding them. Since most of us can’t really handle AI, and it won’t do without a crash course in some pretty advanced math, we’ll focus on metaphors and analogies.
The feedback pointed out some literary devices that can help readers understand the AI phenomenon.
First someone walking by hikikomorphism on Bluesky suggested a phrase “a hungry ghost trapped in a glass” as a guide to whether you are using AI sensibly. He says that if you can replace “hungry ghost trapped in a jar” with “AI” in your description of what you’re doing and it still makes sense, you’re probably using AI in a believable way.
“Take ‘I’ve got a lot of hungry ghosts in jars, they mostly write SQL queries for me’. Sure. Reasonable use case,” writes hikikomorphism. “My girlfriend is a hungry ghost I trapped in a jar”? No. Confused.”
Second, we’re now confronted with endless AI-written content we didn’t ask for: fake romance novels, AI search digests, AI dating digests, AI anything. We need a way to summarize our response to these texts.
Well, one of the most popular acronyms of the Internet era is “tl;dr,” which means “too long, I haven’t read.” Therefore new phrase “ai;dr”the meaning of which should be clear from the context.
Finally, Feedback is flooded with anecdotes of people using AI for important tasks, only to screw it up in spectacular fashion. You might have seen the one where the venture capitalist is asking for an AI tool to arrange the work area on his wife’s computer only to say “ooops” because he deleted 15 years worth of photos (he later got them back).
Or the one where AI hallucinates three months of analytics data.
With these stories in mind, we’ll give the final word to writer Nick Pettigrew. Huh wrote on Bluesky: “I’m convinced that artificial intelligence is the radio of our generation—a discovery with truly useful applications in specific, controlled conditions that we foolishly put into everything from children’s toys to toothpaste until we realized the damage too late, when future generations will wonder if we’ve gone crazy.”
The feedback said more about it, but our AI deleted it – a phrase sure to become the new “the dog ate my homework”.
Cue bits
Somehow, feedback got around all these years without learning about the existence of a quantum information theorist Toby Cubitt.
Do you have a story for feedback?
You can send stories to Feedback by email at feedback@newscientist.com. Please provide your home address. You can see feedback from this week and previous ones on our website.
Leave a Reply