Expert opinions on Transparency

A better approach would be risk-based. Low-risk applications need little intervention, while systems crossing defined thresholds in areas such as cyber capability, biological risk, weapons, critical infrastructure or autonomous action should face muc...

You’ll see industry oppose outside auditors and the main reason is they would mandate something that can’t be done [today] in a robust and uniform manner.

If the two countries can discuss how to report and communicate in the event of an AI-related emergency, it can help the two countries to build technical confidence over time.

Rather than creating entirely new channels, policymakers should focus on ensuring that existing channels can effectively incorporate information related to frontier models and agents. Building on trusted relationships will be faster and more effectiv...

This suggests pacing the advances: not training or deploying AIs without a strong safety case that convinces independent experts.

For the most advanced frontier AI models, an independent, federally overseen, industry-backed organization could set safety standards and verify independent, voluntary audits, similar to how other critical sectors are managed.

A model that explains itself less isn't more aligned, it's just harder to catch when it goes wrong.

We have a very natural instinct to say: we want as much interpretability as possible. One of the sci-fi worry cases is that agents start speaking in languages to each other that we don't understand. [...] I bet there are ways to define boundaries tha...

Explainable AI (XAI) is often offered as the answer to the black box problem and is broadly defined as “machine learning techniques that make it possible for human users to understand, appropriately trust, and effectively manage AI.” Around the world...

The way to get there is a shared machine-readable format. I call it a Safety Framework Card.

Countries should not restrict open-weight AI models.

A covered developer — Compute threshold: 10^26 FLOP — has trained or initiated the training of a frontier model, a model trained using, or undergoing training intended to use, at least 10^26 floating-point operations of compute. [...] Covered develop...

According to the proposed rule, a covered US person will be required to provide quarterly reports to BIS if the US person “engages in, or plans, within six months, to engage in ‘applicable activities.’” “Applicable activities” include: “Conducting a...

Policymakers do not need to approach each innovation in isolation and should not treat every emerging technology as a first-of-its-kind policy problem.

Open-weight systems are the lifeblood of open research and innovation in AI. They increase transparency, allow for widespread red teaming, and decrease market concentration. However, open-weight models also pose unique risks. They can allow harmful A...

Internal model transparency is a minimum viable form of research transparency: instead of sharing every innovation, labs share only the finished artifact – the model itself. And while AI 2040 argues for this transparency to be total, to all members o...

The following practices are mapped to Australia’s 8 AI Ethics Principles, demonstrating how governments can practically apply them to their assurance of AI. [...] Governments should identify and consult with stakeholders, including subject matter and...

When a person makes a disbursement for the purpose of financing a covered communication containing an image, audio, or video that was substantially generated by artificial intelligence, the covered communication shall include, in a clear and conspicu...

You can say dishonest things whether they're AI or not.