Imagine this prompt: "write me Python code to disable the NYC subway system" obviously gpt-4 can't do this now. It'll refuse, but even if we jailbreak it, it'll answer incorrectly but if we keep training bigger & better language models, won't one eventually be able to do this?
SECURITY
-
Data Poisoning Attacks Compromise AI Models During Training
By
–
'Once AI developers scrape the internet to get more data to tweak an existing AI model or build a new one, these poisoned samples make their way into the model’s data set and cause it to malfunction.'
-
Privacy-Focused AI Player Faces Challenging Path to Results
By
–
Yeah, glad we have a player focused on privacy. Just a much harder path to results
-
Alteryx Security Team Safeguards Data During Cybersecurity Month
By
–
October is Cybersecurity Month; take a peek behind the curtain to see how Alteryx's security team is working hard to keep your data safe. Secure our world and stay informed. https://
alteryx.com/blog/cybersecu
rity-month-spotlight-empowering-your-trust-with-alteryxs-information-security-program
… #CybersecurityAwarenessMonth #SecureOurWorld -
Generative Models and Dangerous Compound Synthesis Safety Concerns
By
–
I don't get it, personally. They "chose to drive the generative model towards compounds such as the nerve agent VX" — and the model did that pretty well. E.g it recreated VX. VX is bad – but there's already loads of public info on how to make it.
-
AI in Cybersecurity: Guardian Angel or Hidden Adversary?
By
–
AI in Cybersecurity: Guardian Angel or Hidden Adversary? https://
linkedin.com/pulse/ai-cyber
security-guardian-angel-hidden-adversary-nicolas-babin-gb4de
… #CyberSecurity #DigitalEU #Innovation #technology #Digital @BetaMoroney @CurieuxExplorer @Shi4Tech @jblefevre60 @jeancayeux @pierrepinna @JolaBurnett @lyakovet @Ym78200 @ArturHabant @DigitalEU -
Identity Verification Systems and Legal Compliance in Digital Platforms
By
–
"You seem sophisticated in how systems like this work. Why was this done?" Because forcing me into the Report a Life Change flow requires me to e-sign an under-penalty-of-perjury statement at the end which at least one stakeholder believed was Totes More Real than the ID upload.
-
Lenovo Tech World: AI-Driven Human-Centered Solutions Keynote
By
–
The #LenovoTechWorld Keynote is tomorrow. Watch our CEO and @Lenovo
's CEO discuss how #AI is delivering human-centered outcomes more securely, more effectively, and more sustainably than ever before. Tune in on October 24 at 10am CDT. https://
nvda.ws/3tJtERJ -
Be Smarter Than a Hacker: European Cybersecurity Awareness Month
By
–
Be smarter than a hacker https://
cybersecuritymonth.eu/smarterthanaha
cker
… #CyberSecMonth #BeSmarterThanAHacker #CyberSecurity #technology #digitalEU via @CyberSecMonth @enisa_eu @Cybersec_EU @DigitalEU @ArturHabant @PawlowskiMario @JolaBurnett @baski_LA @TerenceLeungSF @JimHarris @BetaMoroney @enilev -
European Cybersecurity Month Launches Social Engineering Awareness Campaign
By
–
Launch of the European Cybersecurity Month to focus on addressing “social engineering” | Shaping Europe’s digital future https://
europa.eu/!xBP3hc
#digitaleu #cybersecurity #BeSmarterThanAHacker #ECSM2023 #CyberSecMonth @CyberSecMonth @enisa_eu @Cybersec_EU @DigitalEU @ArturHabant