Leaning on a chatbot makes you worse at spotting fake news

MIT's Media Lab ran a four-week test of whether chatbots actually make people better at telling real news from fake. The answer cuts both ways — and the downside lingers after the chatbot is gone.

People who used an AI assistant to check news got better at spotting fakes while they used it — and measurably worse once they stopped. In the study, 67 participants judged whether pairs of headlines and images were real over four weeks: with chatbot help their accuracy on suspect items rose 21%, but by week four, unaided accuracy was 15% below where it started — and roughly a quarter of them walked away convinced they had improved. The work, led by Pattie Maes' Fluid Interfaces group at the MIT Media Lab with Anku Rani and Valdemar Danry as co-leads, was published at CHI 2026 under a title that says the quiet part out loud: dialogues with AI reduce beliefs in misinformation but build no lasting discernment skills.

The trap is that the assistance genuinely works in the moment. The paper found the chatbot cut false beliefs about fake headlines rather than just hedging — the problem is what it left behind. Like GPS atrophying our navigation sense, offloading judgment to a model means the underlying skill never gets exercised. And because the tool felt helpful, most users never noticed the drift: their confidence moved in the opposite direction from their actual ability. That gap is the dangerous part for a platform ecosystem already drowning in synthetic media.

There's a design lever hiding in the data. Assistants that simply handed over answers fostered dependence, while those that coached through Socratic questioning built more durable judgment in users — at the cost of being slower and more annoying to use. The researchers frame it as a speed-versus-effort trade-off, which is exactly the trade-off consumer AI products have no incentive to make. A chatbot that quizzes you before confirming a fact loses to one that just answers, even if the second one leaves you sharper.

Our take: this is a product-design problem as much as a research finding. The same industry shipping AI summaries atop every search result has little reason to ship the "ask, don't tell" variant — so expect dependency, not discernment, to remain the default outcome unless regulators or platforms start treating cognitive offloading as a cost worth pricing in.

What to watch: whether any mainstream assistant ships a "coaching mode" for news evaluation — and whether anyone uses it.

If your main news habit is asking a chatbot what's real, has your own BS detector gotten sharper or rustier? Tell us in the comments.

AI 101 — How do AI video generators work?

An AI video generator is a model that starts with a grid of random noise and repeatedly refines it into a moving picture that matches your prompt — the same trick image generators use, stretched across time. Nothing is filmed, and nothing is stitched together from clips the model has seen. The pixels in the finished video were invented, frame by frame, all at once. Why it matters right now Kling 4.0, Kuaishou's flagship video model, went fully live on October 15 with up to 30 seconds of vid

Houmo moves compute inside memory — and A20 enters production

Beijing chipmaker Houmo used its partner conference this week to lay out two compute-in-memory routes at once — and to say the first chip built on the newer one is already in production. Houmo's A20, its first chip built on a 3D compute-in-memory architecture, has entered production. The part supports up to 16 chips fully interconnected, claims a 50% improvement in compute utilization over the previous generation, and covers models from 3B to 30B parameters, according to co-founder and product

Kling 4.0 goes live with 30-second native video at 10-bit HDR

Kuaishou's Kling AI has spent two years becoming the video model people actually finish projects with. Its new flagship, Kling 4.0, went fully live on October 15 — and this round is less about prettier frames than about the two things that make a generated shot usable in a real edit: length and control. Kling 4.0 generates up to 30 seconds in a single native pass, outputs 10-bit HDR at 4K, 1080p and 21:9, and takes up to 15 reference assets in one task. That budget splits across ten images, fi

Brain organoids classified Braille at 83% — in a trio, not alone

Two stories today: human brain tissue on a chip decoded a tactile signal, and the data-centre company behind some of the largest AI contracts in Europe turns out to be six months behind on its flagship site. Human forebrain organoids grown on a microelectrode array classified Braille letters with 61 percent accuracy on their own and 83 percent when the responses of three organoids were combined, in a peer-reviewed study from the University of Bristol. The team took event-based tactile readings

Read more

AI 101 — How do AI video generators work?

Houmo moves compute inside memory — and A20 enters production

Kling 4.0 goes live with 30-second native video at 10-bit HDR

Brain organoids classified Braille at 83% — in a trio, not alone