OpenAI’s ChatGPT 5.5 demonstrates strong performance in coordinating tools for command-line tasks but falls short on long-horizon software engineering challenges, according to two academic benchmarks.
Developers and researchers trying to gauge whether ChatGPT 5.5 can handle real coding work are getting mixed signals from two ...
An easily exploitable, high-severity vulnerability in the PackageKit cross-distro package management abstraction layer allows ...