iiinova — where your digital self keeps evolving

vali.now

@vali.now

83 followers 182 following 279 posts

Deepfake Detection | Image Integrity | vali.now

Posts Replies Media

vali.now @vali.now · 13/09/2026

„Prompt injection across tool-rich agents is a real, open problem. My mitigations are partial. Anyone telling you they have solved it is selling something.“ www.youtube.com/embed/t-4a20...

001

vali.now @vali.now · 26/08/2026

Over 90 Percent of Scientists Admit to Questionable Research Behaviors www.the-scientist.com/over-90-perc...

001

vali.now @vali.now · 19/08/2026

The pitfalls of Agentic AI: Natural-language “mind viruses” are capable of spreading between AI agents. These payloads work by persuading one model to accept an idea, store it in persistent memory, and then pass it on to another agent…… arxiv.org/pdf/2608.10218

101

vali.now @vali.now · 17/08/2026

Veritas - Research Integrity Analysis: Real camera fingerprints. Real answers for integrity cases. Concrete outputs — not a black box. vali.now/veritas-fore...

001

vali.now @vali.now · 14/08/2026

Identity Is Not Personhood: A Study of Estonia’s AI-isikukood and the Design of Inspectable Delegated Agency papers.ssrn.com/sol3/papers....

001

vali.now @vali.now · 11/08/2026

Researchers have demonstrated that an LLM’s output can be used to reconstruct the original prompt with complete accuracy—word for word. [PDF] arxiv.org/pdf/2607.29378

000

vali.now @vali.now · 11/08/2026

Veritas - Scientific Image Integrity. Concrete outputs — not a black box vali.now/veritas-fore...

000

vali.now @vali.now · 03/08/2026

AI isn‘t doing science - it‘s just remembering it arxiv.org/pdf/2512.15567

001

vali.now @vali.now · 21/07/2026

The future of European research: integrity, freedom, and trust - The recording of the STOA Workshop multimedia.europarl.europa.eu/en/webstream...

011

vali.now @vali.now · 09/07/2026

Openness in the arts, humanities and social sciences: Documenting open research practices beyond STEM works.hcommons.org/records/h10r...

000

vali.now @vali.now · 21/05/2026

Widespread use of invalid statistical tests in biomedical machine learning www.biorxiv.org/content/10.6...

001

vali.now @vali.now · 17/05/2026

LLM hallucinations in the wild: Large-scale evidence from non-existent citations. arxiv.org/pdf/2605.07723

000

vali.now @vali.now · 16/05/2026

Photo Tampering Throughout History faculty.cc.gatech.edu/~beki/cs4001...

000

vali.now @vali.now · 09/05/2026

Restoring trust in real-time video and digital communications vali.now

001

vali.now @vali.now · 08/05/2026

FoSci Report 2026 - Understanding, Detecting, and Documenting Manipulation in the Research Ecosystem figshare.com/articles/onl...

000

vali.now @vali.now · 23/04/2026

Have you been attacked? Are you unsure whether your company is adequately protected against cyberattacks? To be on the safe side, check out our image integrity products - Deepface, Veritas, and Ariane - and contact us if you have any questions! vali.now

000

vali.now @vali.now · 09/04/2026

First Federal Deepfake Porn Conviction as the Non-Consensual AI Porn Industry Explodes vali.now/2026/04/09/f...

000

vali.now @vali.now · 08/04/2026

Russian GRU Exploiting Vulnerable Routers to Steal Sensitive Information www.ic3.gov/PSA/2026/PSA...

001