PulseAugur
EN
LIVE 21:40:06
Polski(PL) Gość zainfekował pytanie egzaminacyjne ukrytą instrukcją dla AI… Dzisiejszym bohaterem dr. Jason Gibson, profesor historii na Alcorn State University (USA). Ale

AI models used in exam and to find critical software vulnerability

By PulseAugur Editorial · [2 sources] ·

A history professor at Alcorn State University, Dr. Jason Gibson, discovered that a hidden instruction within an exam question led 32 out of 35 students to include nonsensical information about Madagascar in their assignments. Separately, a researcher reportedly used a subscription to GPT-5.6 (Sol) to find a critical vulnerability in WordPress Core, potentially worth millions, which allows unauthorized server access. AI

IMPACT Demonstrates AI's potential for both academic disruption and security research, highlighting the need for robust AI safety and verification.

RANK_REASON The cluster describes the use of AI models for unexpected purposes: influencing exam answers and finding software vulnerabilities, rather than a direct release or research breakthrough.

Read on Mastodon — fosstodon.org →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

AI models used in exam and to find critical software vulnerability

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster describes the use of AI models for unexpected purposes: influencing exam answers and finding software vulnerabilities, rather than a direct release or research breakthrough.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
product, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
60 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [2]

  1. Mastodon — fosstodon.org TIER_1 Polski(PL) · [email protected] ·

    Guest infected exam question with hidden instruction for AI... Today's hero is Dr. Jason Gibson, a history professor at Alcorn State University (USA). But

    Gość zainfekował pytanie egzaminacyjne ukrytą instrukcją dla AI… Dzisiejszym bohaterem dr. Jason Gibson, profesor historii na Alcorn State University (USA). Ale przejdźmy od razu do rzeczy. Gdzieś w środku normalnego zadania egzaminacyjnego pojawiły się białe litery na białym tle…

    LINKS sekurak.pl/gosc-zainfekowal-pytanie-egzam…
  2. Mastodon — fosstodon.org TIER_1 Polski(PL) · [email protected] ·

    Spent ~100 PLN on a GPT 5.6 (Sol) subscription. He found an exploit worth ~2,000,000 PLN. Honestly, I didn't believe such things could happen. Researcher,

    Wydał ~100 zł na subskrypcję GPT 5.6 (Sol). Ten znalazł mu exploita wartego ~2000000zł Szczerze mówiąc nie wierzyłem, że takie rzeczy się mogą wydarzyć. Badacz, który niedawno zgłosił krytyczną podatność w WordPress Core, opracował całość z wykorzystaniem narzędzia od OpenAI. Pod…

    LINKS sekurak.pl/wydal-100-zl-na-subskrypcje-gp…

RELATED ENTITIES

RELATED TOPICS