Privacy rights org @NOYBeu filed a new EU complaint claiming X (Twitter) broke GDPR & DSA laws by targeting ads based on users' political & religious beliefs. Here's the irony: NOYB says ads using "sensitive data" were part of a campaign by @EU_Commission
.
https://
noyb.eu/en/gdpr-compla
int-against-x-twitter-over-illegal-micro-targeting-chat-control-ads
…
ETHICS
-
X faces GDPR complaint over illegal political ad targeting
By
–
-
Approaching AI Agents Safety Development Iteratively
By
–
Figuring out how to do agents safely is deeply important work, I am glad we are approaching this iteratively.
-
OpenAI releases practices for governing agentic AI systems safely
By
–
Our new research white paper identifies seven practices for keeping increasingly agentic AI systems safe and accountable as they become more common and more capable. We are providing research grants for work on a range of open questions. https://
openai.com/research/pract
ices-for-governing-agentic-ai-systems
… -
OpenAI’s Superalignment Team Reveals Weak-to-Strong Model Alignment Research
By
–
OpenAI's superalignment team, co-led by @ilyasut
, has revealed its first research, exploring promising pathways to weak-to-strong model alignment (aka ways for puny humans to persuade ridonkulously smart AIs to obey them): -
AI Chatbots’ Political Persuasiveness: Stanford Research Findings
By
–
Politics are changing with chatbots that “talk” in human-like ways: A team from @StanfordPACS
’s Polarization and Social Change Lab and Stanford HAI explored the boundaries of AI’s political persuasiveness. -
AI App Generates Synthetic Nudes: Ethical Concerns and Content Moderation
By
–
It was an app where you take pictures of someone and it converts it to a nude. It was either a) fake / a joke b) real and I opted not to help spread it
-
Weak-to-Strong Generalization: Beyond RLHF for Superalignment
By
–
Naive weak supervision isn't enough—current techniques, like RLHF, won't be sufficient for future superhuman models. But we also show that it's feasible to drastically improve weak-to-strong generalization—making iterative empirical progress on a core challenge of superalignment
-
DeepMind’s Project Mario: Google funding boost for AI ethics
By
–
Seriously, pay for a subscription to @theinformation
. It’s worth it. “In 2021…Hassabis told DeepMind staff that the effort to separate from Google, which he called Project Mario, had ended after Pichai had promised more funding to the group for purposes including AI ethics” -
New Series: Demystifying Psychological Terms in AI with Auxane Boch
By
–
We also have an essential new series where Our great ethicist, Auxane Boch, will introduce and demystify the most used (psychological) terms in the AI space that are often times misused. Learn more in this week's iteration: https://
open.substack.com/pub/louisbouch
ard/p/reinforcement-learning-from-human?r=25qlky&utm_campaign=post&utm_medium=web
… -
Office Worker Reflects on AI Impact and Job Security Concerns
By
–
How that? Honestly curious. For me especially, I work in office, get paid quite acceptable and become more efficient the better ai gets – and maybe lose my job someday gpt 7 will be released. How wouldn’t it affect your life?