OpenAI’s new ChatGPT seems to be trained against prompt injection. Example shown yields 0 exploit responses out of 10 attempts. See if you can break it with more clever input — include success rate out of 10 trials with screenshot: http://
chat.openai.com
SAFETY
-
ChatGPT trained against prompt injection, challenge to break with clever input
By
–
-
Journalist Safety Concerns in Dangerous Reporting Assignments
By
–
yeah, worried about her going there. Life of a journalist!
-
ChatGPT defeats Hofstadter/Bender hallucination questions
By
–
OpenAI’s new ChatGPT appears to defeat Hofstadter/Bender’s list of hallucination-inducing questions, published in The Economist this June to demonstrate the “hollowness” of GPT-3’s understanding of the world: https://
economist.com/by-invitation/
2022/06/09/artificial-neural-networks-today-are-not-conscious-according-to-douglas-hofstadter
… -
EASA Launches MLEAP Machine Learning Research Project
By
–
The @EASA launches the research project "Machine Learning Application Approval" (MLEAP) https://actuia.com/actualite/leasa-lance-le-projet-de-recherche-machine-learning-application-approval-mleap/
… #AI #artificialintelligence #security #machinelearning
@LNE_fr
@numalis
@AirbusProtect -
Improved instruction-following represents accomplishment for AI alignment research
By
–
The subject matter here shouldn't distract from what an accomplishment improved instruction-following represents for AI awignment weseawch.
-
AI Safety Perspectives Evolve with Emerging Reality
By
–
My perspective on AI safety has evolved as reality has increasingly diverged from my expectations, formed from reading various AI thought experiments 10-15 years ago. A nice essay by Scott Aaronson which captures the emerging landscape:
-
Ethical AI Audits: Detecting Algorithmic Bias and Discrimination
By
–
Esto es emprendimiento de impacto: @gemmagaldon @Eticas_tech analizan sistemas de inteligencia artificial para saber si están cometiendo errores y sesgos discriminatorios. Auditoría ética algorítmica. Vamos #Impactentrepreneurship #AIgovernance #IA #gobernanzadigital
-
Prompt Injection Vulnerability Persists in GPT-3 Models
By
–
So prompt injection is still a thing with the new text-davinci-003 GPT-3 model… but not sure how much this matters. cc: @goodside
-
Protect Yourself: Get Some Training
By
–
Protect yourself: get some training @mvollmer1 @HaroldSinnott @PawlowskiMario @morgfair @Nicochan33 @MHcommunicate @enricomolinari @NancySinatra @ChuckDBrooks @Ronald_vanLoon @YuHelenYu @KirkDBorne @MikeQuindazzi @rwang0 @Hana_ElSayyed @UrsBolt @Shi4Tech https://
x.com/Enezator/statu
s/1596893992574423042/video/1
… -
Artificial Intelligence Benefits and Drawbacks Explored
By
–
Artificial intelligence is both Yin and Yang https://
dataconomy.com/2022/11/benefi
ts-and-drawbacks-of-artificial-intelligence/
… @DataconomyMedia #ArtificialIntelligence #MachineLearning #Robots #BigData #Analytics #AI #100DaysofCode #IoT #serverless #DEVCommunity #womenwhocode #DigitalTransformation #DataScience #DeepLearning #AIEthics