To ensure that AI systems remain safe as they start to exceed human capabilities, we’ll need to develop techniques for scalable oversight: the problem of supervising systems’ behavior without assuming that the overseer understands the task better than the system being trained.
REGULATION
-
Equality, Equity and Justice in AI: What Are Your Thoughts?
By
–
About #equality, #equity and #justice What do you think? #Inspiration
-
China’s Data Advantage and Global AI Diversity Challenges
By
–
3) China has a massive data advantage with 1Bn+ people but much of the data lacks diversity (racial /cultural) for a global audience. TikTok bring an exception but then again it’s data practices should be looked at carefully @vonderleyen
-
Meta’s Ad Policy Rejection: License Verification Gone Wrong
By
–
FB rejected my license so I can't run policy ads after I was verified for years. FB then said go to a notary and fill out their form. So I did that. FB rejected my FB form because it wasn't the FB form despite it being the FB form. Welcome to advertising on Meta in 2022.
-
Federated Social Networks: Reconsidering the Cost-Benefit Analysis
By
–
I thought that the massive usability cost and governance complexity of a federated social network wasn't worth it… until a narcissist billionaire bought my favourite online hangout on a whim, and now the costs of federation are looking pretty good to me!
-
Big Tech’s Climate Impact: COP27 and AI Solutions
By
–
The interesting dilemma of #tech & the #ClimateCrisis – does big tech help solve or contribute to problems with the global climate crisis ? Big Tech’s seat at COP27 https://
buff.ly/3zNlg3B #COP27 #ai #artificialintelligence #MachineLearning #deeplearning -
Trust Breach: Reconsidering AI Ethics and Corporate Accountability
By
–
A week ago I would have agreed with you, but this feels different now: for me the trust thermocline here has been well and truly broken
-
U.S. Banks Flag Over $1 Billion in Ransomware Transactions 2021
By
–
U.S. banks flagged #ransomware related transactions adding up to more than $1 billion in 2021, the Treasury Department said. But the true numbers may actually be higher, says Paul Benda
@ABABankers
#cybersecurity #hacking #ciso https://
wsj.com/articles/repor
ted-ransomware-incidents-costs-soared-in-2021-treasury-says-11667513649
… -
Content Moderation: Survey on Human-AI Partnership
By
–
According to a survey, humans and AI should be combined for effective online content moderation https://actuia.com/actualite/selon-un-sondage-humains-et-ia-doivent-etre-associes-pour-une-moderation-de-contenu-en-ligne-efficace/
… #AI #artificialintelligence