AI - AI Feta, the news about scientific AI research

tekoäly

Tekoälystä on enemmän hyötyä testien kirjoittajana kuin koodivelhona

Kaikille on käynyt niin: puhelinsovellus päivittyy, ja jokin tuttu toiminto lakkaa toimimasta. Taustalla on usein harmitonkin muutos, joka rikkoi jotakin toisaalla. Ohjelmistot kasvavat kerros kerrokselta, ja niiden siivoaminen – rivien järjestely, nimenten yhdenmukaistaminen ja vanhan monimutkaisuuden purkaminen – pelottaa, koska pienikin virhe voi näkyä käyttäjälle. Perinteinen vastalääke on kirjoittaa testejä, jotka varmistavat,

AI

Safe Answers Can Still Teach Risky Skills, Study Finds

Even when advanced AI systems refuse to give dangerous instructions, their seemingly harmless answers can be reused to teach smaller models risky skills. A new study shows that safety filters at the output level are not enough on their own. This matters because it affects how quickly powerful know‑how

AI

Graph neural networks can act as fast problem‑solving shortcuts

Cornell University researchers report that a type of AI called a graph neural network can learn to solve classic routing puzzles on its own and produce answers in one shot. This matters because many real tasks — from delivery planning to chip design — boil down to such puzzles, where speed and

AI

Making AI steadier at reading emotions in mental‑health texts

Researchers have built a method to make artificial intelligence more reliable when it reads emotions in text, such as clinical notes, counselling chats and posts in online support groups. This matters because early triage and risk assessment often depend on what people write and how that writing is interpreted. Why

AI

How language models can learn without new labels

Researchers have a new explanation for why some AI systems get better without human feedback. The work matters because it may cut the need for expensive labeled data, but also because self-improving systems can lock in their own mistakes if left unchecked. Why this is being discussed now A team

AI

A modular way to adapt AI models: transplanting trained “organs” between systems

A new study proposes a practical way to adapt large language models by moving trained slices of them between compatible models—much like transplanting an organ. The method promises better accuracy with less training and lets organisations share expertise without handing over their original data. If it works broadly, updating

AI

TruthTensor tests AI in motion: real‑world choices, not quiz questions

A new study proposes a clearer way to test artificial intelligence when the world does not stand still. Instead of quiz-like benchmarks, the approach follows live events and asks models to assign probabilities, then tracks how accurate, stable and careful they are over time. This matters because many organisations now

fraud

Phone scams follow a script — this study teaches a computer to spot it mid-call

Researchers in South Korea have built a system that helps people recognise phone scams as they unfold. The tool, called ScriptMind, uses a text‑generating program to track the steps of a scam conversation and warn the user in real time. This matters because fraudsters now run long, personalised calls

AI

Study: Chatbots can be talked into wrong answers — and fast

Large language models, the systems behind today’s chatbots, can be pushed to change their answers after just a few lines of persuasive prompting. A new research paper finds that smaller models give in especially quickly, and that a common safety trick — asking the model to state how confident it

AI

A new, broader test shows what AI data assistants can and cannot do

Researchers have built a large, practical test for AI tools that help with data work. The results suggest these systems handle routine, well-structured tasks fairly well, but still stumble when the data are messy or when images and text must be understood together. This matters because many organisations are starting

AI

Chatbots should fit human thinking, not expect perfection

Conversational AI is fast becoming a starting point for questions, plans and everyday decisions. A new paper argues that these systems should be built for how people actually think, not for an ideal user who never gets tired, rushed or distracted. The goal is simple: better decisions, fewer avoidable mistakes.

AI

When wording steers the judge: AI evaluations can flip with small changes

Small changes in wording can sway how today’s text-based AI systems judge answers, a new study finds. This matters because such models are increasingly used to grade work, screen content and compare other AI systems. If the question is phrased differently, the verdict can shift. Why this is being