Google has confirmed it will sign the European Union’s new AI Code of Practice, despite voicing concerns about its impact.
The voluntary code, released earlier this month, aims to guide providers of general-purpose AI (GPAI) on transparency, copyright, and safety. Signing it offers companies more legal certainty under the EU’s stricter AI Act, which comes into effect on 2 August.
Kent Walker, President of Global Affairs at Google’s parent company Alphabet, said the code is “closer to supporting Europe’s innovation goals.” However, he warned that some elements could hurt the continent’s competitiveness.
“Departures from EU copyright law or rules exposing trade secrets may slow approvals and chill AI model development in Europe,” Walker wrote.
Although the code is not mandatory, companies that refuse to sign—like Meta—may face tighter inspections. Meta claimed the rules stifle innovation and said it would not join the list of signatories.
The EU’s AI Act gives companies with existing tools two years to comply. Any new tools released after 2 August must follow the rules immediately.
The European Commission will publish a full list of companies that have agreed to the code on 1 August.
Google said it remains committed to working with the EU’s AI Office to keep the code in line with fast-changing AI technology.
The code’s creation has sparked debate, with rightsholders raising alarms about potential copyright violations.
Despite reservations, Google’s move marks a key step in aligning US tech with Europe’s AI governance.
Ask ChatGPT
