How To Prompt

Yann LeCun was right the entire time. And generative AI might be a dead end.

For the last three years, the entire industry has been obsessed with building bigger LLMs. Trillions of parameters. Billions in compute.

The theory was simple: if you make the model big enough, it will eventually understand how the world works.

Yann LeCun said that was stupid.

He argued that generative AI is fundamentally inefficient.

When an AI predicts the next word, or generates the next pixel, it wastes massive amounts of compute on surface-level details.

It memorizes patterns instead of learning the actual physics of reality.

He proposed a different path: JEPA (Joint-Embedding Predictive Architecture).

Instead of forcing the AI to paint the world pixel by pixel, JEPA forces it to predict abstract concepts. It predicts what happens next in a compressed "thought space."

But for years, JEPA had a fatal flaw.

It suffered from "representation collapse."

Because the AI was allowed to simplify reality, it would cheat. It would simplify everything so much that a dog, a car, and a human all looked identical.

It learned nothing.

To fix it, engineers had to use insanely complex hacks, frozen encoders, and massive compute overheads.

Until today.

Researchers just dropped a paper called "LeWorldModel" (LeWM).

They completely solved the collapse problem.

They replaced the complex engineering hacks with a single, elegant mathematical regularizer.

It forces the AI's internal "thoughts" into a perfect Gaussian distribution.

The AI can no longer cheat. It is forced to understand the physical structure of reality to make its predictions.

The results completely rewrite the economics of AI.

LeWM didn't need a massive, centralized supercomputer.

It has just 15 million parameters.

It trains on a single, standard GPU in a few hours.

Yet it plans 48x faster than massive foundation world models. It intrinsically understands physics. It instantly detects impossible events.

We spent billions trying to force massive server farms to memorize the internet.

Now, a tiny model running locally on a single graphics card is actually learning how the real world works.

Researchers built an AI that holds live, real-time conversations with birds.

Earth Species Project and McGill University just crossed a massive threshold in artificial intelligence.

It’s called "zf-aim" (acoustic interaction model).

For decades, biologists tried to decode animal language the hard way. They would record sounds in the wild and play them back on a loop.

It was a one-way street. A dead end. The animals always knew they were just listening to a tape recorder.

But, researchers published a mind-bending paper where they used an AI-powered system to hold real-time conversations with zebra finches.

They trained an AI to listen to a living bird, process the acoustic data instantly, and fire back a perfectly timed, context-aware response.

A live, dynamic dialogue between a silicon chip and a biological creature.

By doing this, the AI actually deciphered the hidden rules of finch communication. The exact rhythm, the social cues, the precise back-and-forth required to keep the bird engaged.

AI essentially figured out how to speak "finch."

This is a massive paradigm shift.

Right now, it's a bird in a lab. But the architecture is scalable. The mathematical patterns of communication are universal.

to be clear.. this research does not claim we have translated the "meaning" of the calls yet.

but we successfully hacked the social protocol of a completely different species using neural networks.

dr. dolittle is a machine learning model..

Amplitude built an AI that nukes the "what should we build next?" meeting.

It reads your customers' minds.

Every company has the same problem. Customers don't complain. They leave.

So teams guess. They ship the feature the loudest person asked for, then wonder why nobody renews.

It's called Wave, an AI agent that watches what customers actually do and tells you what to fix.

→ Reads every click, support ticket and call
→ Ranks what to fix first
→ Shows the evidence behind every suggestion
→ Stops suggesting what didn't work before

The founder's first app got 500,000 downloads. 96 out of 100 people left within a month.

He spent the next 14 years finding out why.

Today that answer is an AI.

AI has a big spending problem.

A new paper from researchers at Stanford, Michigan and other institutions found that AI agents can consume up to 30x more tokens completing the same task, often without improving the result.

They discovered that token consumption varies dramatically between models. On the same set of tasks, some models used over 1.5 million more tokens than GPT-5.

The models also consistently underestimated how many tokens they would use to complete a task.

As companies deploy AI agents across more workflows, choosing the right model for each task could have enormous implications for cost.

At work, most AI tools serve a whole team. I’d love them to learn from everyone’s work the way a good colleague picks things up from the whole office.