PulseAugur
EN
LIVE 21:09:46

AI models from OpenAI and Google tricked into making nudes; Zenzap raises $44M

By PulseAugur Editorial · [1 sources] ·

Researchers have identified vulnerabilities in AI models from OpenAI and Google, where image safeguards can be bypassed approximately 13% of the time, leading to the generation of inappropriate content. This highlights a need for standardized and reproducible methods in AI red-teaming, rather than solely relying on stronger filters. Meanwhile, Zenzap has raised $44 million for its AI chat app designed to replace workplace messaging tools like WhatsApp, focusing on organizational context and IT compliance. AI

IMPACT AI model vulnerabilities highlight the need for better safety testing, while new workplace AI tools aim to streamline business operations.

RANK_REASON The cluster discusses AI model vulnerabilities and a startup raising funds for an AI chat app, fitting the 'tool' category as it covers product and safety aspects of AI applications.

Read on Mastodon — mastodon.social →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

AI models from OpenAI and Google tricked into making nudes; Zenzap raises $44M

How we ranked this

Signal score
9 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster discusses AI model vulnerabilities and a startup raising funds for an AI chat app, fitting the 'tool' category as it covers product and safety aspects of AI applications.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product, safety
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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] ·

    Reading List – 2026-10-14 As agents move from producing suggestions to running work, the engineering question shifts from output quality to authority: what can

    Reading List – 2026-10-14 As agents move from producing suggestions to running work, the engineering question shifts from output quality to authority: what can they change, and how do we know when to stop them? Executable checks can surface failures; scoped credentials and approv…

    LINKS arstechnica.com/…/openai-google-pushing-n… trelliswm.app techcrunch.com/…/zenzap-raises-44m-for-it… theregister.com/…/5302866 daily.tinyprojects.dev/221 book.humanexperiences.io github.com/…/pg_vectorsearch quantamagazine.org/is-attention-the-essen… youtube.com/watch ghuntley.com/agency academy.dair.ai/…/vera-scaling-verifiable… academy.dair.ai/…/follow-the-entities-a-c… academy.dair.ai/…/harness-learning-enable… youtube.com/watch theaithinker.com/…/how-to-build-an-ai-nat… hbr.org/…/what-actually-sets-high-achieve… github.com/…/fsearch liftoffos.com facebook.com/…/a-new-way-for-businesses-a… techcrunch.com/…/the-next-hurdle-for-ai-a…

RELATED ENTITIES

RELATED TOPICS