PSA: Everything you are writing online anywhere on any subject is being used as training data for #GPT4. So are the books you write and youtube videos and podcasts you record. We are all training #GPT now. #ChatGPT #GPT3 #openai #artificialinteligence #infosec #ooda
CYBERSECURITY
-
Stability SDK failure dumps all data
By
–
Okay looks like if Stability SDK fails it dumps all of this
-
Accidental passenger data exposure incident on flight
By
–
Oh crap, I just doxxed everyone else on the flight!
-
Avoid Computer Updates: Security Risk Warning
By
–
Easy, don’t update your computer pic.twitter.com/rBD5QETOWc
— David Herrmann (@herrmanndigital) 16 décembre 2022Easy, don’t update your computer
-
Kingston IronKey Security Solutions Praised
By
–
Always a pleasure, thanks so much, S #KingstonIsWithYou #IronKey #Security #CyberSecurity
-
DDoS Threats Rising in North America: New Security Research
By
–
State of #DDoS in North #America New #Infographic & #research @NETSCOUT Evolving #Risk & #Security support https://
bit.ly/DDoSRise #CyberSecurity #partner #innovation #data #USA @mikeflache @GFCYBER #Analytics @Paula_Piccard @xlth #100DaysOfCode @RLDI_Lamy @andi_staub -
Five Important IT and Security Trends for 2023
By
–
Five Important #IT And #Security #Trends For 2023 Although some of the worst disruption caused by the global Covid-19 pandemic is hopefully behind us, #2023 is shaping up to be another challenging year for #business and #society. https://
linkedin.com/pulse/five-imp
ortant-security-trends-2023-bernard-marr
… -
Privacy in AI: Beyond Training and Model Usage
By
–
And finally, privacy is… hard! While a lot of work focuses on training and using models privately, this is a narrow view of privacy, which encapsulates much more. 14/n
-
FHE Limitations for Large Model Inference Privacy
By
–
But FHE is not so practical for these models. And it may be infeasible to use some of these large models on a user's device. So even with private fine-tuning, then privacy at inference time still remains. 11/n
-
Privacy in Machine Learning: Scaling Challenges and Solutions
By
–
3. Scale makes ML hard to use in a truly private fashion. If you want to do inference on a point without sharing it, you either have to fine-tune&run the model locally (see e.g. this nice paper by @simran_s_arora @HazyResearch https://
arxiv.org/abs/2205.13722), or use FHE encryption. 10/n