
AI coding helpers get FAILing grade
Purdue researchers expose generative AI tools like Copilot's frequent errors when asked basic development questions.

Richi Jennings is a former developer and marketer. He’s also written or edited for Computerworld, Microsoft, Cisco, Micro Focus, HashiCorp, Ferris Research, Osterman Research, DevOps.com, Orthogonal Thinking, Native Trust, Elgan Media, Petri, Cyren, Agari, Webroot, HP, HPE, NetApp on Forbes and CIO.com. Bizarrely, his ridiculous work has even won awards from the American Society of Business Publication Editors, ABM/Jesse H. Neal, and B2B Magazine.
find Richi Jennings on:

Purdue researchers expose generative AI tools like Copilot's frequent errors when asked basic development questions.

Deep learning model knows what keys you press — “with 95% accuracy.” The password's days are numbered.

Your app sec team should factor in more capable malicious AI tools, coming soon.

It’s an optional trial program (for now). How would your devs cope?

The big-fish IAMaaS cloud identity service provider opens its kimono. What can you learn from the exposure?

The privacy of user data is one thing — but security of that data is equally important.

C’mon, dev teams — it's about time to get serious about memory safety, XSS and SQLi.

It’s a dog-eat-dog world ... Bug allows bad actor to manipulate URLs and extract data. Note to devs: Avoid consecutive object references and add entropy.

Forward-thinking DevOps shops are doing it already. Isn’t it time your team got on board?
Get your 14-day free trial of Spectra Assure
Get Free TrialMore about Spectra Assure Free Trial