alignment.org

The Alignment Research Center (ARC) is a non-profit research organization whose mission is to align future machine learning systems with human interests.

We are currently pursuing theoretical research on how to produce formal mechanistic explanations of neural network behavior.

## Recent research

In 2025, ARC has been making conceptual and theoretical progress at the fastest pace that I've seen since I first interned in 2022. Most of this progress has come about because
…
[»](https://www.alignment.org/blog/competing-with-sampling/)

Over the last few months, ARC has released a number of pieces of research. While some of these can be independently motivated, there is also a more unified research vision behind them. The
…
[»](https://www.alignment.org/blog/a-birds-eye-view-of-arcs-research/)

ARC's current research focus can be thought of as trying to combine mechanistic interpretability and formal verification. If we had a deep understanding of what was going on inside a neural
…
[»](https://www.alignment.org/blog/formal-verification-heuristic-explanations-and-surprise-accounting/)

[Read more →](https://www.alignment.org/blog)

## About ARC

**What is “alignment”?** ML systems can exhibit goal-directed behavior, but it is difficult to understand or control what they are “trying” to do. Powerful models could cause harm if they were trying to manipulate and deceive humans. The goal of [intent alignment](https://ai-alignment.com/clarifying-ai-alignment-cec47cd69dd6?ref=alignment-research-center-2.ghost.io) is to instead train these models to be helpful and honest.

**Motivation**: We expect that modern ML techniques would lead to severe misalignment if scaled up to large enough computers and datasets. Practitioners may be able to adapt before these failures have catastrophic consequences, but we could reduce the risk by adopting scalable methods further in advance.

**What we’re working on**: We're currently working on [outperforming random sampling](https://www.alignment.org/blog/competing-with-sampling/) when it comes to understanding neural network outputs. More broadly, we are trying to produce [formal mechanistic explanations](https://www.alignment.org/blog/formal-verification-heuristic-explanations-and-surprise-accounting/) for neural network behaviors in order to [produce robustly aligned systems](https://www.alignment.org/blog/a-birds-eye-view-of-arcs-research/). We see this as the most promising approach to our broader research agenda, which is explained along with our research methodology in our report on [Eliciting Latent Knowledge](https://docs.google.com/document/d/1WwsnJQstPq91_Yh-Ch2XRL8H_EpsnjrC1dwZXR37PC8/edit).

**Methodology**: We’re unsatisfied with an algorithm if we can see any plausible story about how it eventually breaks down, which means that we can rule out most algorithms on paper without ever implementing them. The cost of this approach is that it may completely miss strategies that exploit important structure in real

... (truncated, 4 KB total)