Anthropic has launched Cyber Mission, a major cybersecurity initiative using its powerful Claude AI models to identify and fix security vulnerabilities across critical infrastructure and open-source software.
The announcement comes amid growing concerns over AI-driven cyberattacks, with Anthropic now looking to combat threats its own industry has helped accelerate. And with Salesforce recently deepening its partnership with Anthropic through Claudeforce, the move could have wider implications for the CRM giant’s AI ambitions.
What We Know About Cyber Mission
Anthropic’s Cyber Mission will initially focus on two key areas – protecting critical infrastructure and securing open-source software.
Through its new Critical Infrastructure Defense Program, Anthropic will work with major technology and cybersecurity companies, including Accenture, CrowdStrike, and Palo Alto Networks, to identify and fix vulnerabilities in essential systems such as power grids and water supplies.
It has also launched OSS Scanner, a free tool that uses Claude to regularly scan open-source software for vulnerabilities and potential fixes.
“Highly cyber-capable AI models are widely available to attackers now. But defensive tools – including our own – have not yet reached enough of the defenders who need them,” Anthropic said in a statement announcing the new tools, adding that, despite best efforts, many areas of technology remain “dangerously exposed”.
However, there is an interesting irony amid this. AI companies like Anthropic have spent years developing powerful models, and those same capabilities are now making it easier for cybercriminals to identify and exploit weaknesses in computer systems.
Anthropic acknowledges this problem, warning that AI has reduced the cost of exploiting vulnerabilities while fixing them remains a slow, largely human-led process. The company believes AI could give defenders the upper hand within two years, although it admits that may not be the case in the short term.
In some ways, Anthropic is positioning itself as part of the solution to a problem its own industry has helped accelerate.
Moreover, the announcement is particularly interesting given Salesforce’s expanding relationship with Anthropic. Through Claudeforce, the two companies are bringing Claude further into Salesforce’s enterprise AI strategy, including Agentforce and Slack.
As Salesforce encourages businesses to give AI agents greater autonomy and access to sensitive company information, questions around security and trust are more important than ever. Anthropic’s latest investment in cybersecurity could therefore benefit Salesforce in the longer term, particularly as more businesses need reassurance that powerful AI systems can be used and deployed safely.
Final Thoughts
At the time of writing, there has been no announcement that the new Cyber Mission tools will be integrated into Salesforce or Agentforce.
For now, the question is still whether the companies racing to make AI so powerful can also keep up with the growing security risks that come with it.
