Christian Scharrer, Enterprise Architect und CTO Ambassador bei Dell Technologies in Deutschland, warnt vor den zunehmenden Cyberangriffen auf KI-Systeme basierend auf manipulierten Eingaben oder kompromittierten Trainingsdaten. #CyberSecurity #Cyberthreats #DataPoisoning
www.sysbus.eu/?p=30224
www.sysbus.eu/?p=30224
November 7, 2025 at 5:31 AM
Christian Scharrer, Enterprise Architect und CTO Ambassador bei Dell Technologies in Deutschland, warnt vor den zunehmenden Cyberangriffen auf KI-Systeme basierend auf manipulierten Eingaben oder kompromittierten Trainingsdaten. #CyberSecurity #Cyberthreats #DataPoisoning
www.sysbus.eu/?p=30224
www.sysbus.eu/?p=30224
The TechBeat: The Illusion of Scale: Why LLMs Are Vulnerable to Data Poisoning, Regardless of Size (11/3/2025) #Technology #Cybersecurity #DataPoisoning #MachineLearning #CyberThreats
The TechBeat: The Illusion of Scale: Why LLMs Are Vulnerable to Data Poisoning, Regardless of Size (11/3/2025)
How are you, hacker? 🪐Want to know what's trending right now?: The Techbeat by HackerNoon has got you covered with fresh content from our trending stories of the day! Set email preference here. ##...
puretech.news
November 3, 2025 at 1:03 PM
The TechBeat: The Illusion of Scale: Why LLMs Are Vulnerable to Data Poisoning, Regardless of Size (11/3/2025) #Technology #Cybersecurity #DataPoisoning #MachineLearning #CyberThreats
The Ouroboros Of Machine Learning
When machines start eating their own vomit
indi.ca
November 2, 2025 at 11:49 PM
I endorse this idea. #datapoisoning
I've been doing it for years. Or have I? But then so have Google themselves. We all know you can't trust Google search, but every search you try, including the failed ones, is faithfully tracked and recorded. They've already set up a piss loop of their own!
I've been doing it for years. Or have I? But then so have Google themselves. We all know you can't trust Google search, but every search you try, including the failed ones, is faithfully tracked and recorded. They've already set up a piss loop of their own!
October 29, 2025 at 11:20 AM
I endorse this idea. #datapoisoning
I've been doing it for years. Or have I? But then so have Google themselves. We all know you can't trust Google search, but every search you try, including the failed ones, is faithfully tracked and recorded. They've already set up a piss loop of their own!
I've been doing it for years. Or have I? But then so have Google themselves. We all know you can't trust Google search, but every search you try, including the failed ones, is faithfully tracked and recorded. They've already set up a piss loop of their own!
Can 250 files poison a massive AI? Learn what backdoors are, why they matter, and how to defend.
#AIPoisoning #AI #security #BackdoorAttacks #UCL #CyberDefense #AIthreats #DataPoisoning #StayAhead
Read this and stay a step ahead. www.freeastroscience.com/2025/10/are-...
#AIPoisoning #AI #security #BackdoorAttacks #UCL #CyberDefense #AIthreats #DataPoisoning #StayAhead
Read this and stay a step ahead. www.freeastroscience.com/2025/10/are-...
Are AI Models Easy to Poison? The New Evidence, Explained
Can 250 files poison a massive AI? Learn what backdoors are, why they matter, and how to defend. Read this and stay a step ahead.
www.freeastroscience.com
October 21, 2025 at 6:20 PM
Can 250 files poison a massive AI? Learn what backdoors are, why they matter, and how to defend.
#AIPoisoning #AI #security #BackdoorAttacks #UCL #CyberDefense #AIthreats #DataPoisoning #StayAhead
Read this and stay a step ahead. www.freeastroscience.com/2025/10/are-...
#AIPoisoning #AI #security #BackdoorAttacks #UCL #CyberDefense #AIthreats #DataPoisoning #StayAhead
Read this and stay a step ahead. www.freeastroscience.com/2025/10/are-...
The Illusion of Scale: Why LLMs Are Vulnerable to Data Poisoning, Regardless of Size #Technology #Cybersecurity #DataPoisoning #LLMs #CyberSecurity
The Illusion of Scale: Why LLMs Are Vulnerable to Data Poisoning, Regardless of Size
We stand at an inflection point in AI, where Large Language Models (LLMs) are scaling rapidly, increasingly integrating into sensitive enterprise applications, and relying on massive, often untrusted,...
puretech.news
October 19, 2025 at 1:45 AM
The Illusion of Scale: Why LLMs Are Vulnerable to Data Poisoning, Regardless of Size #Technology #Cybersecurity #DataPoisoning #LLMs #CyberSecurity
Researchers Find It's Shockingly Easy to Cause AI to Lose Its Mind by Posting Poisoned Documents Online futurism.com/artific... #AI #cybersecurity #datapoisoning #poisoned #documents #posted #online
October 15, 2025 at 2:34 PM
Researchers Find It's Shockingly Easy to Cause AI to Lose Its Mind by Posting Poisoned Documents Online futurism.com/artific... #AI #cybersecurity #datapoisoning #poisoned #documents #posted #online
26% of respondents in our State of #InformationSecurity Report 2025 experienced #AI #DataPoisoning in the past year. SC Media UK explores the AI findings from the report, with insight from our CEO, Chris Newton-Smith: https://www.scworld.com/brief/ai-data-poisoning-prevalence-examined
October 14, 2025 at 8:00 AM
26% of respondents in our State of #InformationSecurity Report 2025 experienced #AI #DataPoisoning in the past year. SC Media UK explores the AI findings from the report, with insight from our CEO, Chris Newton-Smith: https://www.scworld.com/brief/ai-data-poisoning-prevalence-examined
Gezielte Data-Poisoning-Angriffe bedrohen KI-Systeme im Finanzsektor: Minimale Manipulation, maximaler Schaden. #Aktuell #Security #Strategie #DataPoisoning #Datensicherheit #KünstlicheIntelligenz #xDSDS25
https://www.it-finanzmagazin.de/data-poisoning-im-finanzsektor-wenn-angreifer-ki-von-innen-her
https://www.it-finanzmagazin.de/data-poisoning-im-finanzsektor-wenn-angreifer-ki-von-innen-her
Data Poisoning im Finanzsektor – Wenn Angreifer KI von innen heraus sabotieren
Ein bislang unterschätztes Risiko bei KI: gezielte Angriffe auf die Datenbasis von KI-Systemen mittels Data Poisoning.
www.it-finanzmagazin.de
October 13, 2025 at 2:52 PM
Gezielte Data-Poisoning-Angriffe bedrohen KI-Systeme im Finanzsektor: Minimale Manipulation, maximaler Schaden. #Aktuell #Security #Strategie #DataPoisoning #Datensicherheit #KünstlicheIntelligenz #xDSDS25
https://www.it-finanzmagazin.de/data-poisoning-im-finanzsektor-wenn-angreifer-ki-von-innen-her
https://www.it-finanzmagazin.de/data-poisoning-im-finanzsektor-wenn-angreifer-ki-von-innen-her
L’enverinament de dades #DataPoisoning és una amenaça real per a la fiabilitat de la #IA.
Al nostre web, Jose Manuel Reche explica com aquests atacs poden comprometre els models i com protegir-los amb estratègies eficaces.
👉 Llegeix l'article: ithinkupc.com/actualitat/q...
#Ciberseguretat
Al nostre web, Jose Manuel Reche explica com aquests atacs poden comprometre els models i com protegir-los amb estratègies eficaces.
👉 Llegeix l'article: ithinkupc.com/actualitat/q...
#Ciberseguretat
Enverinament de dades: el risc ocult que amenaça la IA
Descobreix què és l’enverinament de dades i com posa en risc la seguretat i la fiabilitat dels models d’intel·ligència artificial actuals.
ithinkupc.com
October 13, 2025 at 8:17 AM
L’enverinament de dades #DataPoisoning és una amenaça real per a la fiabilitat de la #IA.
Al nostre web, Jose Manuel Reche explica com aquests atacs poden comprometre els models i com protegir-los amb estratègies eficaces.
👉 Llegeix l'article: ithinkupc.com/actualitat/q...
#Ciberseguretat
Al nostre web, Jose Manuel Reche explica com aquests atacs poden comprometre els models i com protegir-los amb estratègies eficaces.
👉 Llegeix l'article: ithinkupc.com/actualitat/q...
#Ciberseguretat
Full article: www.technadu.com/llm-data-poi...
How can organizations secure their AI models from poisoned training data? Share your perspective below.
#AIsecurity #LLM #Anthropic #CyberSecurity #MachineLearning #DataPoisoning #TechNadu
How can organizations secure their AI models from poisoned training data? Share your perspective below.
#AIsecurity #LLM #Anthropic #CyberSecurity #MachineLearning #DataPoisoning #TechNadu
LLM Data Poisoning Risk: LLMs Can Be Poisoned by Small Samples, Research Shows
New research from Anthropic reveals that a small number of malicious documents can create backdoor vulnerabilities in LLMs, posing significant AI security risks.
www.technadu.com
October 10, 2025 at 1:04 PM
Full article: www.technadu.com/llm-data-poi...
How can organizations secure their AI models from poisoned training data? Share your perspective below.
#AIsecurity #LLM #Anthropic #CyberSecurity #MachineLearning #DataPoisoning #TechNadu
How can organizations secure their AI models from poisoned training data? Share your perspective below.
#AIsecurity #LLM #Anthropic #CyberSecurity #MachineLearning #DataPoisoning #TechNadu
A small number of malicious files - just 250 - can poison large language models, introducing hidden backdoors and manipulated outputs.
Anthropic and UK AI Security Institute research warns this threat scales with model size.
#AIsecurity #LLM #Anthropic #DataPoisoning #CyberSecurity #TechNadu
Anthropic and UK AI Security Institute research warns this threat scales with model size.
#AIsecurity #LLM #Anthropic #DataPoisoning #CyberSecurity #TechNadu
October 10, 2025 at 1:04 PM
A small number of malicious files - just 250 - can poison large language models, introducing hidden backdoors and manipulated outputs.
Anthropic and UK AI Security Institute research warns this threat scales with model size.
#AIsecurity #LLM #Anthropic #DataPoisoning #CyberSecurity #TechNadu
Anthropic and UK AI Security Institute research warns this threat scales with model size.
#AIsecurity #LLM #Anthropic #DataPoisoning #CyberSecurity #TechNadu
Attackers exploit specific trigger phrases or rare token combinations to induce unintended or malicious behaviors in LLMs. This highlights the need for vigilance against subtle, targeted poisoning methods. #DataPoisoning 3/6
October 10, 2025 at 7:00 AM
Attackers exploit specific trigger phrases or rare token combinations to induce unintended or malicious behaviors in LLMs. This highlights the need for vigilance against subtle, targeted poisoning methods. #DataPoisoning 3/6
AI models can acquire backdoors from surprisingly few malicious documents https://arstechni.ca... #UKAISecurityInstitute #alanturinginstitute #AIvulnerabilities #backdoorattacks #machinelearning #datapoisoning #trainingdata #LLMsecurity #modelsafety #pretraining #AIresearch #AIsecurity…
October 9, 2025 at 11:01 PM
AI models can acquire backdoors from surprisingly few malicious documents https://arstechni.ca... #UKAISecurityInstitute #alanturinginstitute #AIvulnerabilities #backdoorattacks #machinelearning #datapoisoning #trainingdata #LLMsecurity #modelsafety #pretraining #AIresearch #AIsecurity…
Poisoned examples flipped sentiment labels in up to 67% of cases, cutting accuracy to about 46.7%. A defense kept accuracy steady, while a regression model reached 100%. https://getnews.me/poisoning-attacks-undermine-in-context-learning-for-health-tweets/ #datapoisoning #healthai
October 7, 2025 at 5:44 PM
Poisoned examples flipped sentiment labels in up to 67% of cases, cutting accuracy to about 46.7%. A defense kept accuracy steady, while a regression model reached 100%. https://getnews.me/poisoning-attacks-undermine-in-context-learning-for-health-tweets/ #datapoisoning #healthai
#AI #DataPoisoning has impacted just over a quarter (26%) of organisations in the UK & US in the last year. Our CEO, Chris Newton-Smith, discusses further findings from our State of Information Security Report: https://www.digit.fyi/a-quarter-of-uk-firms-have-fallen-victim-to-ai-data-poisoning/
October 7, 2025 at 9:00 AM
#AI #DataPoisoning has impacted just over a quarter (26%) of organisations in the UK & US in the last year. Our CEO, Chris Newton-Smith, discusses further findings from our State of Information Security Report: https://www.digit.fyi/a-quarter-of-uk-firms-have-fallen-victim-to-ai-data-poisoning/
One in four (26%) of organisations have been hit by #AI #DataPoisoning in the past year, our State of Information Security Report 2025 found. For CIR Magazine, our CEO, Chris Newton-Smith, shares his thoughts on AI-driven #Cyber threats: https://www.cirmagazine.com/cir/c20250917033.php
October 2, 2025 at 1:01 PM
One in four (26%) of organisations have been hit by #AI #DataPoisoning in the past year, our State of Information Security Report 2025 found. For CIR Magazine, our CEO, Chris Newton-Smith, shares his thoughts on AI-driven #Cyber threats: https://www.cirmagazine.com/cir/c20250917033.php
Trend Micro analizza le minacce agli LLM: da prompt injection a data poisoning, malware embedded e attacchi zero-click, con strategie multilivello di difesa.
#AI #datapoisoning #LLM #OWASP #promptinjection #TrendMicro
www.matricedigitale.it/2025/09/25/i...
#AI #datapoisoning #LLM #OWASP #promptinjection #TrendMicro
www.matricedigitale.it/2025/09/25/i...
September 25, 2025 at 6:28 AM
Trend Micro analizza le minacce agli LLM: da prompt injection a data poisoning, malware embedded e attacchi zero-click, con strategie multilivello di difesa.
#AI #datapoisoning #LLM #OWASP #promptinjection #TrendMicro
www.matricedigitale.it/2025/09/25/i...
#AI #datapoisoning #LLM #OWASP #promptinjection #TrendMicro
www.matricedigitale.it/2025/09/25/i...
Transatlantic EU data transfer: US tech giants strengthened... the #EU is useless, led by Russ phone arseholes from yesteryear.
UTTERLY USELESS THESE UNELECTED #CLOWNS!
#NeverForget! #NeverForgive!
#EncryptEverything
#FuckTheFuckers
#DataPoisoning
#Ooops
#👹
#☠️
UTTERLY USELESS THESE UNELECTED #CLOWNS!
#NeverForget! #NeverForgive!
#EncryptEverything
#FuckTheFuckers
#DataPoisoning
#Ooops
#👹
#☠️
Transferts des données transatlantiques : les géants de la tech américains confortés
Transferts des données transatlantiques : les géants de la tech américains confortés
Le Tribunal de l’Union européenne a rejeté, mercredi 3 septembre, le recours d’un député français demandant l’annulation du Data Privacy Framework, l’accord sur le transfert des données personnelles entre l’Europe et les Etats-Unis.
www.lemonde.fr
September 3, 2025 at 6:14 PM
Transatlantic EU data transfer: US tech giants strengthened... the #EU is useless, led by Russ phone arseholes from yesteryear.
UTTERLY USELESS THESE UNELECTED #CLOWNS!
#NeverForget! #NeverForgive!
#EncryptEverything
#FuckTheFuckers
#DataPoisoning
#Ooops
#👹
#☠️
UTTERLY USELESS THESE UNELECTED #CLOWNS!
#NeverForget! #NeverForgive!
#EncryptEverything
#FuckTheFuckers
#DataPoisoning
#Ooops
#👹
#☠️
Looks like we will join the AI hype train, baby! So to start, let's talk about poisoning AI, and how we can brew something.
jeferson.me/blog/2025/08...
#AI #Bots #Bot #Scraping #DataPoisoning #WebSecurity #Web #Security #Privacy #Hacking #Automation #Exploit #Internet #Freedom #InternetFreedom
jeferson.me/blog/2025/08...
#AI #Bots #Bot #Scraping #DataPoisoning #WebSecurity #Web #Security #Privacy #Hacking #Automation #Exploit #Internet #Freedom #InternetFreedom
Choose your poison
How I poison bots, wreck AI scrapers, and fight back against leeches stealing words without consent.
jeferson.me
August 28, 2025 at 9:44 AM
Looks like we will join the AI hype train, baby! So to start, let's talk about poisoning AI, and how we can brew something.
jeferson.me/blog/2025/08...
#AI #Bots #Bot #Scraping #DataPoisoning #WebSecurity #Web #Security #Privacy #Hacking #Automation #Exploit #Internet #Freedom #InternetFreedom
jeferson.me/blog/2025/08...
#AI #Bots #Bot #Scraping #DataPoisoning #WebSecurity #Web #Security #Privacy #Hacking #Automation #Exploit #Internet #Freedom #InternetFreedom
#datapoisoning CAPTCHAs is a new hobby of mine.
The way it tends to work is that people tend to overmatch rather than under-. So even the tiniest sliver of a bike or motorbike in a square qualifies. But if everyone does that you can correct for it, to a degree.
The way it tends to work is that people tend to overmatch rather than under-. So even the tiniest sliver of a bike or motorbike in a square qualifies. But if everyone does that you can correct for it, to a degree.
August 2, 2025 at 12:26 PM
#datapoisoning CAPTCHAs is a new hobby of mine.
The way it tends to work is that people tend to overmatch rather than under-. So even the tiniest sliver of a bike or motorbike in a square qualifies. But if everyone does that you can correct for it, to a degree.
The way it tends to work is that people tend to overmatch rather than under-. So even the tiniest sliver of a bike or motorbike in a square qualifies. But if everyone does that you can correct for it, to a degree.
Thought-provoking, very relevant article about the shift of the threat landscape to include AI / LLM models trained with 'poisoned data', the impact on risk and security management ...
-
www.csoonline.com/article/4022... #cybersecurity #CISO #ParadigmShift #AI #DataPoisoning
-
www.csoonline.com/article/4022... #cybersecurity #CISO #ParadigmShift #AI #DataPoisoning
AI poisoning and the CISO’s crisis of trust
The CISO’s role has always been to protect the organization from threats it does not yet understand. AI poisoning requires CISOs to rethink risk, architecture, relationships, and shared responsibility...
www.csoonline.com
July 15, 2025 at 7:07 PM
Thought-provoking, very relevant article about the shift of the threat landscape to include AI / LLM models trained with 'poisoned data', the impact on risk and security management ...
-
www.csoonline.com/article/4022... #cybersecurity #CISO #ParadigmShift #AI #DataPoisoning
-
www.csoonline.com/article/4022... #cybersecurity #CISO #ParadigmShift #AI #DataPoisoning
#Datapoisoning represents a powerful addition to the U.S. arsenal of covert capabilities, offering a distinct advantage in the evolving landscape of #AI driven #warfare, argues Aaron Conti. lieber.westpoint.edu/data-poisoni...
Data Poisoning as a Covert Weapon: Securing U.S. Military Superiority in AI-Driven Warfare - Lieber Institute West Point
Data poisoning represents a powerful addition to the U.S. arsenal of covert capabilities, offering a distinct advantage in AI driven warfare.
lieber.westpoint.edu
June 30, 2025 at 1:57 PM
#Datapoisoning represents a powerful addition to the U.S. arsenal of covert capabilities, offering a distinct advantage in the evolving landscape of #AI driven #warfare, argues Aaron Conti. lieber.westpoint.edu/data-poisoni...