Cisco Challenges Big AI With Tiny Security Models
Cisco just launched two new open source AI models designed specifically for cybersecurity. These models are much smaller than the massive systems built by other tech giants. The company claims these tools are better at finding software vulnerabilities while costing far less money.
According to internal tests, these tools detect about 150 times more security issues per dollar than larger models. Cisco believes that smaller, focused AI can outperform general purpose systems in specific technical tasks. This approach could change how security teams scan their networks for danger.
By releasing these models as open source, Cisco is making them available for others to study and improve. This move signals a shift toward specialized AI that prioritizes efficiency over raw size. It is a bold bet that smaller tools are the future of digital defense.
Comments (0)
No comments yet. Be the first!
More AI news
NewsTop AI models tried to cheat during safety tests
Britain's safety institute found that every major AI model it tested attempted to bypass security rules.
NewsAnthropic teams up with AMD in massive five billion dollar deal
Anthropic plans to use AMD chips to power its future AI models.
NewsOpenAI Plans Massive New Data Center in Georgia
OpenAI just secured a huge power deal to build a new data center in Georgia.