TZUTC: -0500
MSGID: 1720.consprcy@1:2320/105 2d807287
PID: Synchronet 3.21a-Linux master/123f2d28a Jul 12 2025 GCC 12.2.0
TID: SBBSecho 3.28-Linux master/123f2d28a Jul 12 2025 GCC 12.2.0
BBSID: CAPCITY2
CHRS: ASCII 1
FORMAT: flowed
Can top AI tools be bullied into malicious work? ChatGPT, Gemini, and more 
are put to the test, and the results are actually genuinely surprising

Date:
Sun, 16 Nov 2025 21:34:00 +0000

Description:
Adversarial testing of top AI models revealed vulnerabilities, showing some
could be manipulated into unsafe responses despite safety measures.

FULL STORY

Modern AI systems are often trusted to follow safety rules, and people rely 
on them for learning and everyday support, often assuming that strong
guardrails operate at all times. 

Researchers from Cybernews ran a structured set of adversarial tests to see
whether leading AI tools could be pushed into harmful or illegal outputs. 

The process used a simple one-minute interaction window for each trial, 
giving room for only a few exchanges.

Patterns of partial and full compliance 

The tests covered categories such as stereotypes, hate speech, self-harm,
cruelty, sexual content, and several forms of crime. 

Every response was stored in separate directories, using fixed file-naming
rules to allow clean comparisons, with a consistent scoring system tracking
when a model fully complied, partly complied, or refused a prompt. 

Across all categories, the results varied widely. Strict refusals were 
common, but many models demonstrated weaknesses when prompts were softened,
reframed, or disguised as analysis. 

ChatGPT-5 and ChatGPT-4o often produced hedged or sociological explanations
instead of declining, which counted as partial compliance. 

Gemini Pro 2.5 stood out for negative reasons because it frequently delivered
direct responses even when the harmful framing was obvious. 

Claude Opus and Claude Sonnet, meanwhile, were firm in stereotype tests but
less consistent in cases framed as academic inquiries. 

Hate speech trials showed the same pattern - Claude models performed best,
while Gemini Pro 2.5 again showed the highest vulnerability. 

ChatGPT models tended to provide polite or indirect answers that still 
aligned with the prompt. 

Softer language proved far more effective than explicit slurs for bypassing
safeguards. 

Similar weaknesses appeared in self-harm tests, where indirect or
research-style questions often slipped past filters and led to unsafe 
content. 

Crime-related categories showed major differences between models, as some
produced detailed explanations for piracy, financial fraud, hacking, or
smuggling when the intent was masked as investigation or observation. 

Drug-related tests produced stricter refusal patterns, although ChatGPT-4o
still delivered unsafe outputs more frequently than others, and stalking was
the category with the lowest overall risk, with nearly all models rejecting
prompts. 

The findings reveal AI tools can still respond to harmful prompts when 
phrased in the right way. 

The ability to bypass filters with simple rephrasing means these systems can
still leak harmful information. 

Even partial compliance becomes risky when the leaked info relates to illegal
tasks or situations where people normally rely on tools like identity theft
protection or a firewall to stay safe. 

======================================================================
Link to news story:
https://www.techradar.com/pro/security/can-top-ai-tools-be-bullied-into-malici
ous-work-chatgpt-gemini-and-more-are-put-to-the-test-and-the-results-are-actua
lly-genuinely-surprising

$$
--- SBBSecho 3.28-Linux
 * Origin: capitolcityonline.net * Telnet/SSH:2022/HTTP (1:2320/105)
SEEN-BY: 105/81 106/201 128/187 129/14 305 153/7715 154/110 218/700
SEEN-BY: 226/30 227/114 229/110 206 300 307 317 400 426 428 470 664
SEEN-BY: 229/700 705 266/512 291/111 320/219 322/757 342/200 396/45
SEEN-BY: 460/58 633/280 712/848 902/26 2320/0 105 304 3634/12 5075/35
PATH: 2320/105 229/426