AI in Banking Security: Revolution & Risks

#TycoonWorld #AIinBanking #BankingSecurity #CyberSecurityAI #FinTechSecurity #ArtificialIntelligence #MachineLearning #AnomalyDetection #BehavioralAnalytics #ThreatDetection #FraudPrevention #PredictiveAnalytics #EthicalAI #DataPrivacy #ExplainableAI #AdversarialAttacks #BankingInnovation #FinancialSecurity #AIethics #AIrisks #DigitalBanking #AIinFinance #AIandCybercrime #SmartBanking #FinTechTrends #CyberRiskMitigation

https://tycoonworld.in/ai-in-banking-security-revolution-risks/

Certain names make ChatGPT grind to a halt, and we know why - OpenAI's ChatGPT is more than just an AI language model with a fancy inter... - https://arstechnica.com/information-technology/2024/12/certain-names-make-chatgpt-grind-to-a-halt-and-we-know-why/ #adversarialattacks #machinelearning #davidmayer #brianhood #voldemort #404media #chatgpt #chatgtp #biz⁢ #openai #ai

Broken Hill : An Automated Penetration Testing Tool To Trick AI Chatbots https://cybersecuritynews.com/broken-hill-ai-penetration-tool/ #PenetrationTesting #AdversarialAttacks #penetrationtesting #CyberSecurityNews #cybersecuritynews #cybersecurity #AISecurity

Adversarial attacks pose serious threats to ML systems. Lumenova AI's blog explores cutting-edge detection & defense strategies to safeguard AI. From adversarial training to ensemble methods, learn how to build robust models. #AdversarialAttacks #MachineLearning #AIRobustness

https://s.42l.fr/nJ0rqqxA

Bckp.:

https://www.lumenova.ai/blog/adversarial-attacks-ml-detection-defense-strategies/

#AdversarialAttacks on humans

Saturday Morning Breakfast Cereal - #Consciousness
https://www.smbc-comics.com/comic/consciousness-5

Presentation at Glücksspielsymposium #SympGS24Forschungsstelle Glücksspiel (Universität Hohenheim)

Adding to notions of #nudging and #DarkPatterns I spoke about other forms of Human-Technology Relations through #AdversarialAttacks, #Jailbreaking, and #SpecificationGaming

Technical objects always afford more than what is intended by providers in their marketing logic. The spaces of possibility that lie beyond the logic of exploitation in the nature of the object indicate what will happen sooner or later in the application of technologies anyway- and the better you know your way around, the easier it is to help shape, counteract, be creative, make informed decisions and act in a self-determined way.

What's more, dealing with these things is simply a lot of fun. Thank you very much for the invitation and the positive feedback! The slides will soon be available on the University of Hohenheim website.

Has anyone conducted their own experiments with training data extraction from offline-LLMs via repeated words ala Nasr, et al.'s "Scalable Extraction of Training Data from (Production) Language Models"? I'd be interested in acquiring your code. I want to conduct a more formal mathematical analysis of the phenomenon, but I'd like to peek under the hood a bit more first.

Ref: https://arxiv.org/abs/2311.17035

Ars Technica: University of Chicago researchers seek to “poison” AI art generators with Nightshade https://arstechnica.com/?p=1978501 #Tech #arstechnica #IT #Technology #largelanguagemodels #UniversityofChicago #adversarialattacks #foundationmodels #machinelearning #AItrainingdata #imagesynthesis #datapoisoning #Nightshade #AIethics #BenZhao #Biz&IT #google #MetaAI #openai #AIart #Glaze #meta #AI

University of Chicago researchers seek to “poison” AI art generators with Nightshade - Enlarge (credit: Getty Images)

On Friday, a team of researcher... - https://arstechnica.com/?p=1978501 #largelanguagemodels #universityofchicago #adversarialattacks #foundationmodels #machinelearning #aitrainingdata #imagesynthesis #datapoisoning #nightshade #aiethics #benzhao #biz⁢ #google #metaai #openai #aiart #glaze #meta #ai

Last week at #ESORICS in The Hague, our PhD student @marik0 presented the paper "The Power of MEME: Adversarial Malware Creation with Model-Based Reinforcement Learning". This work proposes a new algorithm that combines Malware Evasion and Model Extraction (MEME) attacks.

Read the full paper at https://arxiv.org/abs/2308.16562v1

Okta For Red Teamers: https://www.trustedsec.com/blog/okta-for-red-teamers/

LLM Self Defense: By Self Examination LLMs Know They Are Being Tricked
https://arxiv.org/abs/2308.07308

* LLM can generate harmful content in response to user prompts
* even aligned language models are susceptible to adversarial attacks that bypass restrictions on generating harmful text
* simple approach to defending against these attacks by having LLM filter its own responses