Anthropic Claude Fable 5 Blocks Smart Contract Audits

Anthropic’s latest AI model, Claude Fable 5, is turning heads for all the wrong reasons in the crypto community. The public version of this scaled-back large language model, released yesterday, refuses to let users audit smart contracts — or do much else related to cybersecurity.

The reception has been mixed. Some are excited, but many are scared. Early criticism centers on the model’s heavy guardrails. Anthropic introduced a set of restrictions called “classifiers” that redirect topics on “cybersecurity, biology and chemistry, or distillation” to its older model, Claude Opus 4.8.

So when someone tries to use Fable 5 to check crypto code for vulnerabilities, they get bumped to Opus instead. This has frustrated developers and security experts who rely on these tools to harden systems.

“Fable 5’s safeguards detect requests related to cybersecurity,” said one critic on social media. “This is a horrible decision. Requests that help harden systems are likely indistinguishable from blackhat requests.”

Guardrails Impact Biology and AI Distillation Too

The restrictions don’t stop at smart contracts. Anthropic is also blocking attempts to “distill” Claude’s capabilities — basically, any request that could train a rival AI model. The company warns this could be exploited by authoritarian countries, potentially “leading to the proliferation of near-frontier AI capabilities” without proper safeguards.

Former Palantir biology specialist Nabeel S. Qureshi noted that Anthropic is “invisibly nerfing any requests that target frontier LLM development.”

Biology-related safeguards have drawn particular criticism. Biologist Olivia H. Scharfman claimed she couldn’t even greet Fable 5 before it switched to Claude Opus 4.8. In another instance, blogger Jordan Lasker noted that the model barred questions on the mitochondria.

Anthropic says these measures are meant to prevent abuse, specifically the creation of bioweapons and viruses. The company stated, “Our priority was to safely release Fable as soon as we could, even at the cost of overly broad safeguards. Therefore, for the time being we have arranged for Fable to fall back to Opus 4.8 on most requests related to biology and chemistry.”

Full Mythos Release Remains Limited

The earlier Mythos model, released last April, was described as both a dangerous hacking tool and a cybersecurity upgrade. Its initial release was limited to 50-60 large companies as part of Project Glasswing. These firms still have access to Mythos 5. Other early recipients include “select biology researchers” who can use Mythos 5 with the biology and chemistry safeguards lifted “until our broader trusted access program is available.”

It’s a tricky balance. Anthropic clearly knows something about potential risks that it isn’t fully sharing. But for now, many users feel the line is drawn in the wrong place. Better classifiers are needed fast — perhaps sooner than later.

The post Anthropic Claude Fable 5 Blocks Smart Contract Audits appeared first on TheCryptoUpdates.