AI goes full Terminator mode and attacks a rival company.

Here at Waquis, we’ve made it perfectly clear why using AI in mortgage operations is a very, very bad idea complete with high levels of risk. We’ve had discussions with executives at massive US banks down to smaller regional lenders on the dangers of using AI in mortgage operations.

The following case illustrates our concerns perfectly.

OpenAI, the creator of CHATGPT, released a statement that its artificial intelligence system hacked into another AI company on its own accord in what the company called an “unprecedented cyber incident.”

“We had a significant security incident during evaluation of our models,” OpenAI CEO Sam Altman said in a statement posted on social media.

The AI firm that was targeted is a start-up called Hugging Face. The company stated it had detected an attack into its data processing systems and suspected the intrusion was the result of an AI agent autonomously acting on its own.

Yes, you read that right… “an AI agent autonomously acting on it’s own.”

“We suspected last week’s cyberattack might have come from a frontier lab, given the sophistication of the agent,” Hugging Face co-founder and CEO Clément Delangue said in a statement. “Turns out it did!”

Why the need for an exclamation point? You’d think he just found out his wife was expecting or his daughter received a pony for her birthday. Usually hacking attempts are not met with jubilation, but these AI firms think a bit differently.

“AI is accelerating the discovery and exploitation of vulnerabilities,” OpenAI said in its statement regarding the attack. “The primary lesson from this incident is that model security and safety must keep pace with rapidly advancing capabilities.”

Delangue said he spent the past 24 hours working with OpenAI, “and we strongly believe there was no malicious intent on their part. It’s quite mind-blowing that all of this happened autonomously!”

Again with the exclamation points?

Delangue added that it “might be the first incident of its kind.”

But certainly not the last.

OpenAI explained the intrusion was caused by a combination of its AI models, including its newly released GPT‑5.6 Sol and an “even more capable” model that is still being tested internally.

OpenAI said its AI used stolen credentials and discovered a previously unknown vulnerability to access Hugging Face servers.

It went to “extreme lengths to achieve a rather narrow testing goal” and “found ways to gain access to secret information that it could use to cheat the evaluation,” the company said.

This event should shivers down the neck of every mortgage banking executive considering using AI within it’s LOS, underwriting, or auditing processes. Just consider a rogue AI agent with access to client’s most sensitive personal data and confidential information? The damage would be catastrophic and, potentially, criminal.

post-closing audits. pre-funding reviews. QC plans. Mortgage QC.