Microsoft Security Risk Detection Nears Launch

The cloud-based fuzz testing service for developers uses artificial intelligence techniques to track down software vulnerabilities before they become a problem for users.

Microsoft Security Risk Detection, formerly Project Springfield, will be generally available later this summer, the software maker announced.

The cloud-based fuzz testing service uses artificial intelligence (AI) technologies to track down security vulnerabilities and other bugs in their software before they are discovered by attackers out in the wild. Fuzz testing involves subjecting software to conditions that may trigger a crash or create an opening for malicious hacks. 

Nowadays, it's not unusual for technology companies to apply AI techniques to detect security threats that affect corporate networks and their users.

For example, Barracuda's new Sentinel service uses AI to unmask targeted spear phishing attempts. In June, cybersecurity startup Balbix launched a security breach risk assessment platform that uses machine learning, a subset of AI, to predict the likelihood of a breach and estimate how resistant organizations are to attacks on their IT environments.

With its new Security Risk Detection product, Microsoft is encouraging software developers to squash potential security vulnerabilities before they pose a problem to end-users.

"The Microsoft Security Risk Detection service is unique in that it uses artificial intelligence to ask a series of 'what if' questions to try to root out what might trigger a crash and signal a security concern," stated the company in a July 21 announcement. "Each time it runs, it hones in on the areas that are most critical, looking for vulnerabilities that other tools that don’t take an intelligent approach might miss."

Interested developers can test a preview version of Microsoft Security Risk Detection for Windows or Linux now. Sign-ups are available here.

Microsoft is just one of many IT giants currently exploring the intersection of AI and IT.

In June, Lenovo announced plans to expand beyond its PC roots with an ecosystem of smart devices and services. They include the SmartCast+ smart speaker and projector, the company's answer to Amazon's Echo, and virtual assistant named Cava with face recognition capabilities. Meanwhile, Lenovo's Xiaole AI platform supports the creation of personalized user experiences of improved customer service.

Both Fujitsu and Huawei Technologies are reportedly working on new AI processors, potentially placing them in direct competition with Nvidia, Advanced Micro Devices (AMD), Intel and Google.

Fujitsu's so-called deep learning unit (DLU) chip uses low-precision formats to improve performance and lower energy requirements. Huawei's processor, meanwhile, will feature CPU, GPU and AI-specific components, all within a single piece of silicon.

Last month, Google announced new updates to two of its Cloud Machine Learning Perception services.

Cloud Video Intelligence API (application programming interface), a video analysis service that extracts entities within video content and makes it searchable, is now open to all comers. Google also enabled a new feature that automatically detects adult content.

Meanwhile, the company's Cloud Vision API includes a new web-detection feature that is powered by Google Image Search and helps users find similar images along with their associated metadata. 

Improvements include more accurate Safe Search functionality and updated face detection capabilities that can better identify when a subject expresses sadness, surprise and anger.

Credit: eWeek

No comments:

Powered by Blogger.