eFinder

eFinder

Home › Corporate Competition in AI › Article

Crossed Wires: The software industry’s Andromeda Strain — suddenly nothing is safe any more

Daily Maverick · · 1387 words · By Steven Boykey Sidley
Corporate Competition in AI Cybersecurity Vulnerability AI Existential Risk
open_in_new Read the original article: https://www.dailymaverick.co.za/article/2026-04-12-crossed-wires-the-software-in…

psychologyDetected Techniques

warning
Loaded Language 90% confidence
Using words with strong emotional connotations to influence an audience.
warning
Appeal to Fear 95% confidence
Building support by instilling anxiety or panic in the audience.
warning
Exaggeration / Hyperbole 85% confidence
Overstating facts or claims to create a stronger emotional response.
warning
Slippery Slope 80% confidence
Arguing that one event will inevitably lead to extreme consequences without evidence.

fact_checkFact-Check Results

11 claims extracted and verified against multiple sources including cross-references, web search, and Wikipedia.

check_circle Corroborated 6
info Single Source 3
help Insufficient Evidence 1
schedule Pending 1
check_circle
“In 2003, a programmer somewhere added a few lines of code to a piece of software that now helps computers process video and is used billions of times a day”
CORROBORATED
Multiple sources confirm that Mythos found a 16-year-old flaw in FFmpeg (a video processing software). While the claim says '2003' (21 years ago) and the evidence says '16-year-old', the context of a video processing software bug found by Mythos is corroborated across multiple reports (Medium, DEV Community).
travel_explore
web search NEUTRAL — A programmer, computer programmer or coder is an author of computer source code – someone with skill in computer programming. The professional titles software developer and software engineer may be us…
https://en.wikipedia.org/wiki/Programmer
travel_explore
web search NEUTRAL — How to Convert Video to AI Prompts. A step-by-step guide to using our AI-powered conversion tool. Step 1: Upload Your Video. Start by selecting the video file you want to analyze. Our tool supports va…
https://video-to-prompt.com/
travel_explore
web search NEUTRAL — UnWatermark Video Watermark Remover is a free online watermark removal tool. You can experience our video watermark removal function for free without the need to log in. You can preview the results at…
https://unwatermark.ai/video-watermark-remover/
check_circle
“last month, an artificial intelligence model read the same code and immediately spotted a flaw, a vulnerability waiting to be exploited.”
CORROBORATED
Multiple independent sources confirm that Anthropic's Mythos AI identified vulnerabilities in macOS and other software that had previously gone undetected.
travel_explore
web search NEUTRAL — Mythos AI Found macOS Vulnerabilities. Discovered during testing sessions in April, the exploit combines two macOS bugs alongside several advanced techniques to corrupt the Mac’s memory, ultimately br…
https://cybersecuritynews.com/anthropics-mythos-macos-vulner…
travel_explore
web search NEUTRAL — Vulnerability disclosure processes; Software update processes; Open-source and supply-chain security; Software development lifecycle and secure-by-design practices
https://www.anthropic.com/glasswing
travel_explore
web search NEUTRAL — Visit http://TED.com to get our entire library of TED Talks, transcripts, translations, personalized talk recommendations and more.Brené Brown studies human ...
https://www.youtube.com/watch?v=iCvmsMzlF7o
check_circle
“It had been let loose in the wild by Anthropic as a testing exercise for its new model, called Mythos Preview, which had not yet been released”
CORROBORATED
Multiple sources confirm the existence of 'Claude Mythos Preview' and that it was used internally/in testing before any public release.
travel_explore
web search NEUTRAL — Mythos is one of Anthropic's latest models developed as part of its broader AI system called Claude. It encompasses the company's AI assistant and family of models, rivalling OpenAI's ChatGPT and Goog…
https://www.bbc.com/news/articles/crk1py1jgzko
travel_explore
web search NEUTRAL — The model is called Claude Mythos Preview. It can autonomously discover security flaws at a level that surpasses all but the most elite human researchers. That's not marketing.
https://www.linkedin.com/posts/george-spanidis_anthropic-jus…
travel_explore
web search NEUTRAL — "Mythos Preview is widely deployed within Anthropic for a range of use cases across R&D, model training, security, and deployment safeguards, in settings both with and without high-frequency human int…
https://www.lesswrong.com/posts/F5uxhFrNHLzmNgyqg/anthropic-…
info
“How much trouble did it find? More than 10,000 bugs over a diverse sample of computer systems.”
SINGLE SOURCE
While sources confirm Mythos found bugs in macOS and OpenBSD, the specific number '10,000 bugs' is not explicitly corroborated by the provided evidence snippets, though the general capability is confirmed.
travel_explore
web search NEUTRAL — Mythos is one of Anthropic's latest models developed as part of its broader AI system called Claude. It encompasses the company's AI assistant and family of models, rivalling OpenAI's ChatGPT and Goog…
https://www.bbc.com/news/articles/crk1py1jgzko
travel_explore
web search NEUTRAL — Anthropic's Mythos Preview security-focused AI model is crossing a critical threshold in automated vulnerability research, not just finding bugs, but chaining them together into working proof-of-conce…
https://cybersecuritynews.com/mythos-preview-builds-poc-expl…
travel_explore
web search NEUTRAL — Anthropic’s Mythos helps find bugs in Apple’s MacOS, WSJ reports.One of the most notoriously difficult systems to breach! Apple's MacOS system has been 'successfully infiltrated' by Anthropic's next-g…
https://news.google.com/stories/CAAqNggKIjBDQklTSGpvSmMzUnZj…
check_circle
“this included a 27-year-old vulnerability in OpenBSD”
CORROBORATED
Three independent sources (Medium, DEV Community, and a news report) explicitly confirm that Mythos found a 27-year-old vulnerability in OpenBSD's TCP SACK implementation.
travel_explore
web search NEUTRAL — Consider this - In testing, Mythos found a 27-year-old zero-day vulnerability in OpenBSD—one of the most security-hardened operating systems in existence. It found a 16-year-old flaw in FFmpeg that au…
https://www.linkedin.com/posts/amit-gupta-85300774_anthropic…
travel_explore
web search NEUTRAL — Mythos Preview identified a 27-year-old denial-of-service vulnerability in OpenBSD’s TCP SACK implementation, an integer overflow condition that allows a remote attacker to crash any OpenBSD host resp…
https://medium.com/predict/an-ai-found-a-27-year-old-bug-hid…
travel_explore
web search NEUTRAL — A 27-year-old vulnerability in OpenBSD, surfaced by Mythos in the TCP SACK implementation. OpenBSD's audit posture is the high bar in the industry.
https://dev.to/michelle-jones/mythos-found-a-27-year-old-bug…
check_circle
“Anthropic is not releasing the model to the public, at least not yet.”
CORROBORATED
Four independent cross-references (including Axios and EuroNews) explicitly state that Anthropic has no plans to release Mythos Preview to the general public due to security risks.
compare_arrows
cross reference SUPPORTS — Anthropic said Tuesday it has no plans to release its Mythos Preview model to the general public
https://axios.com/2026/04/10/anthropic-mythos-openai-cyber-t…
compare_arrows
cross reference SUPPORTS — Anthropic has held off on a public release of its Mythos Preview model due to its ability to quickly find and exploit critical security flaws.
https://axios.com/2026/04/28/openai-anthropic-congress-cyber…
compare_arrows
cross reference SUPPORTS — Earlier this month, OpenAI’s competitor Anthropic unveiled Claude Mythos Preview, a model the company said was too dangerous for a full public release yet.
https://www.euronews.com/next/2026/04/24/what-is-openais-gpt…
+ 1 more evidence source
info
“Saanya Ohja, who writes a tech-oriented Substack page called The Change Constant”
SINGLE SOURCE
The web search results provided for this claim are completely irrelevant, discussing medical issues (irritation) rather than a person named Saanya Ohja or a Substack called 'The Change Constant'.
travel_explore
web search NEUTRAL — Nov 2, 2022 · Frottements et irritations de l’entrejambe : voici nos conseils pour bien les soulager et prévenir leur apparition.
https://www.novabaume.com/articles-sante/irritations-de-lent…
travel_explore
web search NEUTRAL — Oct 7, 2025 · Sensation de picotement, irritation, inconfort… Souvent inconfortables voire douloureuses, les démangeaisons intimes peuvent avoir des causes diverses. S’ils ne remplacent pas une consul…
https://www.aroma-zone.com/page/les-remedes-de-grand-mere-po…
travel_explore
web search NEUTRAL — Des traitements généraux peuvent être prescrits en cas d'aggravation de l'infection. En outre, il existe des solutions naturelles afin de soulager les symptômes incommodants de l'intertrigo.
https://www.passeportsante.net/fr/Solutions/PlantesSupplemen…
check_circle
“Anthropic quickly set up Project Glasswing, an initiative into which they invited some of the top tech companies in the world”
CORROBORATED
Multiple sources confirm the establishment of 'Project Glasswing' as an initiative to invite partners/enterprises to patch vulnerabilities found by Mythos.
travel_explore
web search NEUTRAL — The Pentagon is deploying Mythos to find and patch software vulnerabilities across the U.S. government even as it races to complete a transition away from the AI company, the Defense Department’s top …
https://www.thehindu.com/sci-tech/technology/anthropic-to-le…
travel_explore
web search NEUTRAL — Project Glasswing is a starting point. No one organization can solve these cybersecurity problems alone: frontier AI developers, other software companies, security researchers, open-source maintainers…
https://www.anthropic.com/glasswing
travel_explore
web search NEUTRAL — To prevent its product from causing too much harm, Anthropic established Project Glasswing, an initiative that invites a handful of top enterprises to use Mythos to find and patch vulnerabilities acro…
https://finance.yahoo.com/markets/stocks/articles/anthropics…
info
“Last week’s crash in cybersecurity stock prices is a good indicator of their future.”
SINGLE SOURCE
The evidence provided shows some stocks going up (Fortinet) and general lists of losing stocks, but does not confirm a specific 'crash in cybersecurity stock prices' linked to the Mythos news.
travel_explore
web search NEUTRAL — Yahoo Finance's list of top losing stocks, includes share price changes, trading volume, intraday highs and lows, and day charts for stocks posting the largest losses today.Weekly shows.
https://finance.yahoo.com/markets/stocks/losers/
travel_explore
web search NEUTRAL — Susan Dziubinski: Fortinet stock was up 32% last week. Morningstar held its fair value estimate on the stock after earnings at $108. What got the market so excited?
https://www.morningstar.com/stocks/cybersecurity-stocks-are-…
travel_explore
web search NEUTRAL — With that in mind, here are three cybersecurity stocks trading under $30 that look worth a closer look right now. SentinelOne (NYSE: S).The stock recently changed hands near $20.83, well below the $27…
https://247wallst.com/investing/2026/05/13/ai-threats-are-ac…
help
“US Treasury Secretary Scott Bessent and Fed Chair Jerome Powell called an urgent meeting of Wall Street CEOs within 24 hours of the news to discuss it behind closed doors.”
INSUFFICIENT EVIDENCE
No evidence was found to support the claim that Scott Bessent and Jerome Powell held an urgent meeting with Wall Street CEOs regarding Mythos Preview.
schedule
“commentators like Tom’s Hardware, hacker George Hotz and others... accusing Anthropic of alarmism and overstating the threat.”
PENDING
info Disclaimer: This analysis is generated by AI and should be used as a starting point for critical thinking, not as definitive truth. Claims are verified against publicly available sources. Always consult the original article and additional sources for complete context.