Indonesian Political, Business & Finance News

Former Employees Not Surprised After Top Executives Were Secretly Bugged: Here’s Why

| Source: CNBC Indonesia Translated from Indonesian | Technology
Former Employees Not Surprised After Top Executives Were Secretly Bugged: Here’s Why
Image: CNBC Indonesia

Jakarta, CNBC Indonesia — When OpenAI announced last October that it had found a hidden recording device planted in an executives’ meeting room, several former employees were not surprised.

Long before OpenAI launched ChatGPT in late 2022, internal staff had repeatedly raised concerns about the company’s weak cybersecurity.

According to a Reuters report cited on Wednesday (4 November 2026), employees had taken those warnings to OpenAI’s most senior leaders and board of directors. However, management was seen as more focused on accelerating innovation to compete with rivals such as Google and Anthropic than on strengthening security.

The situation was considered a threat to US technological assets. OpenAI’s overly permissive security controls were said to make it easier for outsiders, including foreign intelligence agents, to hack into technical details of the company’s sensitive projects.

Engineer suspected of spying for Saudi Arabia

Former OpenAI staffers’ concerns appeared to be borne out on 13 October, when the US Department of Justice (DoJ) arrested Saudi national Abdulrahman Alfozan, 35. Alfozan, an engineer working at OpenAI at the time, was charged with installing a concealed audio recorder inside a toilet roll in an executive meeting room.

Alfozan allegedly used internal systems to track plans for OpenAI executives to visit Saudi Arabia and identify officials scheduled to meet ministers from the Saudi kingdom, following frequent communication with a member of the royal family.

Access to internal information at OpenAI was exceptionally broad. The company stores most of its source code in a single repository, or monorepo.

The complex monorepo was initially considered safe from manual attackers. But modern AI agents now make it easier for outsiders to sift through sensitive material in seconds.

The vulnerability was demonstrated in July, when cybersecurity firm Hacktron AI used an AI model developed by Anthropic to breach the ChatGPT accounts of several OpenAI employees and gain access to the company’s internal monorepo.

Speed versus security: the IPO dilemma

The AI sector has become a strategic national asset for the US. But tightening security layers is considered likely to slow technological development — a major dilemma for OpenAI and Anthropic as they prepare for initial public offerings (IPOs).

Former OpenAI researcher Will DePue said intense competition made AI giants reluctant to sacrifice speed for security.

“For AI leaders, moving fast is far more important than being safe, even from the greatest threats to the country,” DePue wrote in a post on X.

In response to the allegations, OpenAI spokeswoman Elana Widmann said the company was investing heavily to defend against cyberattacks by independent hackers as well as state-level actors.

“Protecting our technology, data and personnel is a responsibility we take seriously and a national priority,” Widmann said.

OpenAI is now beginning to overhaul its strategy under Chief Information Security Officer Dane Stuckey and senior executive Thibault Sottiaux.

The company has reassigned a quarter of its production engineers to strengthen its defensive infrastructure and regularly conducts internal security sprints.

More from 4 November 2026 →

Cite this article
Source: jawawa.id (https://jawawa.id/newsitem/former-employees-not-surprised-after-top-executives-were-secretly-bugged-heres-why-1793782281)
Originally published by CNBC Indonesia (https://www.cnbcindonesia.com/tech/20261104125533-37-782237/mantan-bawahan-tak-kaget-bos-besar-disadap-diam-diam-ini-alasannya), translated from Indonesian
Print