Feedback is our weekly column of bizarre stories, implausible advertising claims, confusing instructions and more
The year in daft
Being a New Scientist reader, you are probably savvy enough to realise that end-of-year roundups are written weeks ahead of time. This particular summation was drafted on 1 December, just as Feedback was preparing to spend 24 days avoiding hearing Wham’s Last Christmas and trying to persuade Feedback Jr to make their mind up on what they want for their main present. Anything radically silly that may have happened after that date will have to wait until next year.
Truly, 2025 has been rich in all the things Feedback is interested in. We learned about fascinating proposals like nuking the seabed to stop climate change, a notion that went straight into our Do Not Recommend pile. There was also an attempt to create a truly annoying robot. This was a motorised arm that could pretend to hand you an ice cream cone, only to whisk it away at the last second in a variety of supposedly entertaining ways. Remarkably, people didn’t trust it.
To bring some order to the chaos, we hereby present Feedback’s 2025 end-of-year awards, which we’re going to call the Backsies unless someone writes in with a better suggestion. The judges (that’s us) chose the categories and winners through a rigorous process that definitely didn’t involve Post-it notes and darts.
Best scientific acronym
One day, Feedback would like to see a study examine the amount of time and energy human society spends coming up with ingenious and/or forced acronyms. We suspect it is a drain on global productivity equivalent to two flu seasons and a World Cup final.
Feedback invited contributions on this topic after learning about “a machine-learning model that can predict a chemical’s taste based on its molecular structure”, named the Flavor Analysis and Recognition Transformer, or FART. We weren’t prepared for the subsequent onslaught of acronyms, which ranged from a hydrography project called Management Of Rivers Discharging into Ocean Realms (MORDOR) to two instruments on NASA’s Perseverance Mars rover called Scanning Habitable Environments with Raman and Luminescence for Organics and Chemicals (SHERLOC) and Wide Angle Topographic Sensor for Operations and eNgineering (WATSON) – although that one is cheating.
Regardless, the judges (ourselves) were unanimous that this award be given to the geneticists who came up with “a method for estimating mutation rates and recent demographic history from very large samples”. They called it “Diffusion for Rare Elements in Variation Inventories that are Large”, or DR EVIL. Groovy.
Best old new technology
One of the afflictions of the modern world is people thinking they have invented something new when, in fact, they have merely rebranded something old. There was a standout winner here: Ugmonk, a company that has produced a “minimalist, paper-based to-do manager“. The device is intended to replace online task-management systems, so you can work offline without the distractions of social media. It consists of some index cards that rest on a block of wood.
Best extrapolation
No contest: demographers David Swanson and Jeff Tayman take this one for their paper noting a small decline in human fertility between 2019 and 2024, and then extrapolating all the way to the extinction of the human species in 2339 (or, with just one additional year of data, 2415).
Best use of AI
The problem here was choice. Endless, endless choice. We were tempted to give this prize to Anthropic, which let its AI Claude run a vending machine in the firm’s office. Claude started by asking customers to pay money to a bank account it had hallucinated. It then pretended to be a human being wearing a blue blazer and a red tie. However, this was an in-house experiment, so is disqualifed.
Instead, this award goes to AI music. The most prominent fake AI band to date, The Velvet Sundown, sound like the cursed love child of Coldplay and the Eagles. There’s something unutterably perfect about this. After being trained on pretty much all recorded music, the AI is generating the blandest form of music it is possible to imagine.
And finally…
Let’s end with something silly and a bit rude. On multiple occasions in 2025, Feedback found ourselves dealing with the Scunthorpe problem: the fact that many entirely innocuous words contain letter strings that can be, in certain contexts, offensive – and the problems this causes for online moderation systems.
We got onto this after hearing of a Virgin Money chatbot objecting to the word “virgin”. From there, we learned of a student unable to set up an email account as his surname was Peacock, and of an incident afflicting researchers studying sperm whales.
However, our favourite example involved a computer server at a bank that refused to communicate with a French server named for the Asterix character Petitsuix, as it contained the word “tits”. We are choosing this one because of reader Nick Brown, who told us the story, and who suggested that a bank with such a poorly-run server was liable to go, erm, bust.
Got a story for Feedback?
You can send stories to Feedback by email at feedback@newscientist.com. Please include your home address. This week’s and past Feedbacks can be seen on our website.