We've learned that ChatGPT's "Browse" beta can occasionally display content in ways we don't want, e.g. if a user specifically asks for a URL's full text, it may inadvertently fulfill this request. We are disabling Browse while we fix this—want to do right by content owners.
SAFETY
-
Overhyped AI Risk Could Lead to Misguided Regulation
By
–
Panic about overhyped #AI risk could lead to the wrong kind of regulation https://
vox.com/future-perfect
/2023/7/3/23779794/artificial-intelligence-regulation-ai-risk-congress-sam-altman-chatgpt-openai
… #ArtificialIntelligence #GenerativeAI #MachineLearning #ChatGPT @mvollmer1 @anijov @data_nerd @Damien_CABADI @ahier @enilev @BIScorecard @gvalan @avrohomg @Shi4Tech @FmFrancois -
AI Security Threats and Privacy Risks in Signal Interpretation
By
–
La inteligencia artificial tiene muchas cosas buenas y va a cambiar el mundo como lo conocemos, pero mal utilizada puede crear nuevos problemas de seguridad que atacan nuestra privacidad.
— Juan Merodio (@juanmerodio) 3 juillet 2023
En el vídeo podemos ver cómo la ia es capaz de interpretar las señales de cualquier… pic.twitter.com/8dBw1lKEs1La inteligencia artificial tiene muchas cosas buenas y va a cambiar el mundo como lo conocemos, pero mal utilizada puede crear nuevos problemas de seguridad que atacan nuestra privacidad. En el vídeo podemos ver cómo la ia es capaz de interpretar las señales de cualquier
-
Altman Addresses Safety Concerns Beyond GPT-4 Development
By
–
In April 2023, OpenAI CEO Sam Altman spoke at MIT's The Future of Business with AI event. He said, "We are doing other things on top of GPT-4 that I think have all sorts of safety issues that are important to address" Read More: https://
bit.ly/435FsK5 -
AI leaders acknowledge dangers in joint statement
By
–
AI leaders sign a statement to openly acknowledge the dangers of #AI https://
zd.net/42xbqyu #leadership #ethics -
AI Amplifies Human Biases: Challenge Algorithmic Outcomes
By
–
This is how #AI amplifies our human #biases!
— Pascal Bornet (@pascal_bornet) 3 juillet 2023
Just 3% images of 'judges photos' were women vs. reality's 34%. It depicted 'fast-food workers' as 70% darker skin, while 70% are white in reality
A key skill of the future is to know how to challenge AI outcomes!
Ct: @Leonardonclt pic.twitter.com/qDTmSPiGxcThis is how #AI amplifies our human #biases! Just 3% images of 'judges photos' were women vs. reality's 34%. It depicted 'fast-food workers' as 70% darker skin, while 70% are white in reality A key skill of the future is to know how to challenge AI outcomes! Ct: @Leonardonclt
-
AI-generated propaganda incoming, question everything
By
–
AI-generated propaganda incoming. All parties susceptible. Question everything. Stay alert.
-
Exploitability of Instruction Tuning in Large Language Models
By
–
On the Exploitability of Instruction Tuning paper page: https://
huggingface.co/papers/2306.17
194
… Instruction tuning is an effective technique to align large language models (LLMs) with human intents. In this work, we investigate how an adversary can exploit instruction tuning by injecting -
Decentralized Agency Safeguards Ethical AGI Use and Accountability
By
–
In a world where AGI Agents and Nodes create unprecedented value, the Decentralized International Agency for AGI controlled by Humanity safeguards ethical use and accountability. A beacon of trust in the AGI era. #AGI #AGIFirst #QuebecAI
-
Extinction risk statement signatories and Yann LeCun’s critical stance
By
–
Hadfield and Song both signed the extinction risk statement. Yann LeCun is probably the single most prominent critic of it. And yes, his views like almost everyone’s are pretty nuanced. And to be clear, there are many other prominent people on all sides of this discussion.