1/3 The EU #AIAct implementation process has been a hot mess so far
Earlier this week, on 2 February, the AI Act's prohibitions on unacceptably risky #AI systems came into effect.
As we cautioned most recently in our statement with 27 organisations and experts, these prohibitions are already too weak to effectively prioritise #HumanRights over AI companies' profits and protect us from further harms.
But that's not all...
https://edri.org/our-work/commission-guidelines-ai-act-implementation-human-rights-and-justice/
2/3 ... Member States are now taking advantage of these weak bans to push for dangerous laws at a national level. This includes legalising #MassSurveillance through Remote Biometric Identification (RBI) systems
This is happening at the very centre of the EU - Belgium is proposing a full copy paste of the AI Act exceptions to the RBI ban!
This is exactly we warned about when the #AIAct passed - it will lead to legalisation of massive #HumanRights violations https://edri.org/our-work/eu-ai-act-fails-to-set-gold-standard-for-human-rights/
3/3 There's also the procedural mess:
delays from the #AIOffice
national implementation delays
#GPAI code of conduct issues
Commission's interpretative guidelines released AFTER bans already entered into force
no guidelines on the AI system definition yet, which will determine which systems fall under the scope of the bans in the first place.
This is terrible for legal certainty, in addition to the #AIAct already falling short of civil society demands to put people's rights first.