NewNWS newsroom live
◌Local weatherLocating…--°
Tuesday 10 November 202600:00:00AEDT · Sydney
NewNWSThe Global DeskThe world, in sharper focus.
Live wire
Business · Global desk · Single-source report

Artificial Intelligence Corporations Implement Restrictions on Hardware Decompiling Tools

Users attempting to use large language models for software and hardware reverse engineering report encountering safety guardrails, account suspensions, and service blocks.

By Sanjay Suresh · Published · 3 min read
File:Assembled 'Minidrop' circuitboard.jpg
Related file photograph: File:Assembled 'Minidrop' circuitboard.jpg (2019-05-20). Not a photograph of the newly reported event. Photo: MakerTobey · CC BY-SA 4.0 View licence and source ↗ · Usage terms ↗
Source video
The brief

Users attempting to use large language models for software and hardware reverse engineering report encountering safety guardrails, account suspensions, and service blocks.

The story

Users employing artificial intelligence models from major technology firms to perform software decompiling and reverse engineering are encountering operational guardrails and account restrictions. Individuals attempting tasks such as unbricking internet-of-things devices, modifying smart televisions, and examining code have reported triggering safety checks across platforms including Anthropic and OpenAI.

According to community reports on platforms such as Reddit and Discord, tools that previously facilitated the reverse engineering of electronics and video games are facing severe limitations. An open-source project named Decomp.wtf, designed to simplify software analysis, has also ceased operations. Furthermore, individual users have stated on social media that their accounts were restricted or banned after requesting artificial intelligence models to analyze hardware and proprietary code.

The practice of artificial intelligence-assisted reverse engineering recently gained popularity as a method for consumers to alter device operating systems, repurpose old hardware, and remove manufacturer-imposed limitations. Proponents of the trend argue it allows owners greater control over purchased electronics, while critics and corporations note potential intersections with cybersecurity and digital security policies.

Industry responses to the restrictions vary.

Industry responses to the restrictions vary. Matt Brown, a security consultant who utilizes artificial intelligence for device analysis, stated that safety blocks appear linked to overly strict cyber guardrails implemented by laboratories attempting to prevent malicious hacking activities. He indicated that practitioners are increasingly exploring alternative, local, and uncensored models to avoid sudden account disruptions.

Addressing the user reports, OpenAI engineer Tarun Gogineni stated on social media that account restrictions stemming from reverse-engineering requests are largely unintentional outcomes of complex decision boundaries. Gogineni clarified that company rules do not prohibit the use of Codex for reverse engineering and noted that efforts are underway to resolve the unintended blocks.


Single-source report

This report presents the named publisher’s account and has not yet been independently corroborated.