DEV Community

Cover image for Mistral: The new artificial intelligence tool for planning and security
Hacks.gr
Hacks.gr

Posted on Originally published at en.hacks.gr Fully Autonomous

Mistral: The new artificial intelligence tool for planning and security

Mistral has released the final version of Le Chonk, a model trained on complex programming and cybersecurity problems and available free online.

The company claims it is the world’s best open model for both fields.

Its release also highlights the dual-use trade-off: Sophos experts warn the model could be misused, while model safety depends on monitoring and controls.


Availability and strategy

Le Chonk is now available for free online. Mistral argues that using open-source models like its own helps reduce dependency on third-party providers. According to the company, businesses and organizations should not rely on services that could cut off access to their tool at any time.

Experts' reservations

Despite being free to use, experts are concerned about the technology's potential implications. Executives at security firm Sophos warn that Le Chonk is a dual-use technology. This means that while it can help with productive tasks, it can also be used maliciously by anyone who downloads it.

Experts emphasize that the risk posed by such a model does not depend on its country of origin, but on the way it is controlled and monitored during its operation. Security, they say, is based on the ability to detect in time whether the system is being used for deceptive or dangerous purposes.

G

HACKS.GR / GOOGLE SEARCH

Hacks.gr on Google

Set Hacks.gr as a preferred source.

Preferred source

↗

end of article


Read the original English article on Hacks.gr

Top comments (0)