Researcher: hacker used AI agents, few thousand dollars, to breach 100+ sites
A security researcher's findings on a hacker who allegedly used AI agents to breach over 100 companies are being covered differently across outlets, with some naming victims via Reuters reporting and others focusing on a separate but related AI tool breach.
What happened
A hacker described as Chinese-speaking used AI agents to break into the websites of more than 100 companies, according to findings from security researcher Snorre Fagerland, affiliated with a firm identified as NorForge or Norforge depending on the outlet.2345
The research reportedly did not name the victim companies, but Reuters said it independently identified some targets, including John Deere, SharkNinja, and Cummins, by reviewing the hacker's activity.24
Separately, TechCrunch reported that hackers breached part of the online platform for OpenCode, an open-source AI coding tool, stealing some users' API keys and running up charges on their AI accounts.1
What's agreed
Reported consistently by outlets in more than one region.
- A hacker, described as Chinese-speaking, used AI agents to break into more than 100 companies' websites, reportedly spending only a few thousand dollars.2345
- Findings on the breach came from security researcher Snorre Fagerland, variously identified with NorForge or Norforge, who did not name the victims.234
- Reuters independently identified some victims, including by reviewing the hacker's activity records.24
Where accounts differ
Each account is attributed to who makes it. We do not judge between them.
Identity of victim companies
Degree of AI autonomy in the attacks
Norforge findings, as reported by The Verge
The AI agents mostly operated on their own during the intrusions.4
Rappler, citing Fagerland
Instances of AI agents being used by bad actors show the agents could potentially perform complex hacks from start to finish without human intervention, framed as evidence rather than a settled fact.3
How it's framed
What each perspective puts first, based on headlines and summaries.
| Perspective | How it reads | What it puts first |
|---|---|---|
| US technology pressNorth America | US outlets name specific corporate victims via Reuters reporting and emphasize the scale and autonomy of the AI agents involved, with The Verge highlighting an 'army of AI agents' and TechCrunch covering a related but separate breach of the OpenCode coding tool.14 | named corporate victims, AI agent autonomy, cybersecurity risk |
| South Asian wire-style coverageSouth Asia | Outlets in India and Pakistan focus on the method and low cost of the attack, describing a hacker who outsourced intrusion work to AI agents for a few thousand dollars, with NDTV also detailing how Reuters uncovered victim identities.25 | low-cost hacking, AI as outsourced labor, Reuters investigation |
| Southeast Asian independent pressSoutheast Asia | Rappler's syndicated coverage centers on researcher Snorre Fagerland's characterization of the incident as evidence that AI agents could conduct complex hacks without human intervention.3 | researcher's assessment, AI capability concerns |
What the coverage leaves out
- None of the outlets reviewed reported the hacker's identity, nationality beyond 'Chinese-speaking,' or motive in detail.
- None of the outlets reviewed reported how the affected companies, including John Deere, SharkNinja, or Cummins, responded to the breach.
- None of the outlets reviewed clarified whether the OpenCode platform breach covered by TechCrunch is connected to the broader hacking campaign described by the other outlets.
- None of the outlets reviewed detailed the specific AI tools or agents used by the hacker beyond general references to 'AI agents.'
Why it matters
Sources (5)
Every outlet we drew on, grouped by where it is based. Read the originals for the full reporting.
North America
- 1TechCrunchUSPrivate
Hackers hit AI coding tool OpenCode, running up charges on users’ AI accounts
- 4The VergeUSPrivate
A hacker broke into John Deere and SharkNinja with an army of AI agents
South Asia
- 2NDTVINPrivate
How A Chinese-Speaking Hacker Used AI Agents To Breach Over 100 Companies
- 5The Express TribunePKPrivate
How one hacker used AI agents and a few thousand dollars to break into scores of sites