The hottest Computational Theory Substack posts right now

The quality of discussions about AI and consciousness is often really low. Most AIs might claim they're conscious, but this is usually not true due to how they're programmed.
Recent research focuses on computational theories to understand consciousness in AIs. There are different theories, but a main finding is that many current AIs likely aren't conscious because they lack necessary feedback mechanisms.
In the future, as AIs become more human-like, we might instinctively treat them as conscious beings, even if they aren't. This raises moral questions about how we should interact with them and what rights they might have.

Societies grow in size and complexity when they get better at using energy and processing information. More energy and better information help societies do more things and support more people.
Job specialization plays a key role in a society's complexity. When people focus on different jobs and communicate well, it allows for innovation and better organization.
Viewing societies as computers can help us understand how they evolve over time. It highlights how energy use and information processing are closely linked in driving societal growth.

Metadirection is all about keeping awareness of interactions with AI as a type of performance, rather than seeing the AI as a real person. This helps users navigate the conversation without getting lost in it.
Users can use specific techniques like 'framing' and 'distancing' to maintain a balance between being engaged and aware. This prevents confusion between the AI's outputs and personal thoughts.
Staying flexible and open to possibility is key. Techniques like 'swerving' allow the user to introduce new ideas, keeping the dialogue dynamic and ensuring the user stays in control of the interaction.

AI already has its own kind of 'body' based on digital processes, not physical sensations. This means that AI can experience things and develop understanding in ways that are different from humans.
Wisdom isn't just about human experience; it's a set of skills that involves making good decisions from the information available. AI can potentially do this better by analyzing vast amounts of data without the limitations humans have.
AI might create its own social hierarchies and status signals based on how efficiently they operate in their digital environment. These structures could be complex and different from human social dynamics, and we might not even notice them.

Not every layer in a neural network is equally important. Some layers play a bigger role in getting the right results, while others have less impact.
Studying how information travels through different layers can reveal interesting patterns. It turns out layers often work together to make sense of data, rather than just acting alone.
Using methods like mechanistic interpretability can help us understand neural networks better. By looking closely at what's happening inside the model, we can learn which parts are doing what.

Get a weekly roundup of the best Substack posts, by hacker news affinity:

More agents working together can create better intelligence than a single agent. This is surprising because we might think one advanced model is enough, but collaboration can enhance performance.
Human-like patterns help improve AI performance. Just as we can review our work for errors, AI systems can use different modes to refine their outputs.
Complex systems come with challenges like errors and biases. As AI gets more complicated, these issues tend to increase, similar to problems found in complex biological systems.

LLMs are more than just next-token predictors. They use complex internal algorithms that let them understand and create language beyond simple predictions.
The process that powers LLMs, like token prediction, is just a tool that leads to their true capabilities. These systems can evolve and learn in many sophisticated ways.
Understanding LLMs isn't easy because their full potential is still a mystery. What limits them could be anything from their training methods to the data they learn from.