Factbox: Governments race to regulate AI tools

Melco International Development

19/09/2023 07:33

(Updates Britain entry with CMA principles, U.S. entry with
Musk comment)
       Sept 19 (Reuters) - Rapid advances in artificial
intelligence (AI) such as Microsoft-backed OpenAI's ChatGPT are
complicating governments' efforts to agree laws governing the
use of the technology.
    Here are the latest steps national and international
governing bodies are taking to regulate AI tools:

    AUSTRALIA
    * Planning regulations
    Australia will make search engines draft new codes to
prevent the sharing of child sexual abuse material created by AI
and the production of deepfake versions of the same material,
the country's internet regulator said on Sept. 8.

    BRITAIN    
    * Planning regulations
    Britain's Competition and Markets Authority (CMA) set out
seven principles on Sept. 18 designed to make developers
accountable, prevent Big Tech tying up the tech in their walled
platforms, and stop anti-competitive conduct like bundling.
    The proposed principles, which come six weeks before Britain
hosts a global AI safety summit, will underpin its approach to
AI when it assumes new powers in the coming months to oversee
digital markets.
    Britain's competition regulator said in May it would start
examining the impact of AI on consumers, businesses and the
economy and whether new controls were needed.

    CHINA
    * Implemented temporary regulations
    China has issued a set of temporary measures effective from
Aug. 15, requiring service providers to submit security
assessments and receive clearance before releasing mass-market
AI products.
    Following government approvals, four Chinese tech firms,
including Baidu  9888.HK  and SenseTime Group  0200.HK ,
launched their AI chatbots to the public on Aug. 31.

    EUROPEAN UNION
    * Planning regulations
    European Commission President Ursula von der Leyen on Sept.
13 called for a global panel to assess the risks and benefits of
AI, similarly to the global IPCC panel which informs policy
makers about the climate.
    EU lawmakers agreed in June to changes in a draft of the
bloc's AI Act. The lawmakers will now have to thrash out details
with EU countries before the draft rules become legislation.
    The biggest issue is expected to be facial recognition and
biometric surveillance where some lawmakers want a total ban
while EU countries want an exception for national security,
defence and military purposes.

    FRANCE
        * Investigating possible breaches
    France's privacy watchdog CNIL said in April it was
investigating several complaints about ChatGPT after the chatbot
was temporarily banned in Italy over a suspected breach of
privacy rules.
    France's National Assembly approved in March the use of AI
video surveillance during the 2024 Paris Olympics, overlooking
warnings from civil rights groups.

    G7
        * Seeking input on regulations
    Group of Seven (G7) leaders meeting in Hiroshima, Japan,
acknowledged in May the need for governance of AI and immersive
technologies and agreed to have ministers discuss the technology
as the "Hiroshima AI process" and report results by the end of
2023.
    G7 nations should adopt "risk-based" regulation on AI, G7
digital ministers said after a meeting in April.

    IRELAND
    * Seeking input on regulations
    Generative AI needs to be regulated, but governing bodies
must work out how to do so properly before rushing into
prohibitions that "really aren't going to stand up", Ireland's
data protection chief said in April.

    ISRAEL
    * Seeking input on regulations   
    Israel has been working on AI regulations to achieve the
right balance between innovation and the preservation of human
rights, Ziv Katzir, director of national AI planning at the
Israel Innovation Authority, said in June.
    Israel published a 115-page draft AI policy in October 2022
and is collating public feedback ahead of a final decision.

    ITALY
        * Investigating possible breaches
    Italy's data protection authority plans to review artificial
intelligence platforms and hire AI experts, a top official said
in May.
    ChatGPT became available to users in Italy in April after
being temporarily banned over concerns by the national data
protection authority in March.

    JAPAN
        * Investigating possible breaches
    Japan expects to introduce by the end of 2023 regulations
that are likely closer to the U.S. attitude than the stringent
ones planned in the EU, an official close to deliberations said
in July.
    The country's privacy watchdog said in June it had warned
OpenAI not to collect sensitive data without people's
permission.

    SPAIN
    * Investigating possible breaches
    Spain's data protection agency said in April it was
launching a preliminary investigation into potential data
breaches by ChatGPT. It has also asked the EU's privacy watchdog
to evaluate privacy concerns surrounding ChatGPT.

    UNITED NATIONS
    * Planning regulations
    The U.N. Security Council held its first formal discussion
on AI in New York in July. The council addressed both military
and non-military applications of AI, which "could have very
serious consequences for global peace and security", U.N.
Secretary-General Antonio Guterres said.
    Guterres in June backed a proposal by some AI executives for
the creation of an AI watchdog like the International Atomic
Energy Agency, but noted that "only member states can create it,
not the Secretariat of the United Nations".
    The U.N. Secretary-General has also announced plans to start
work by the end of the year on a high-level AI advisory body to
review AI governance arrangements.

    U.S.
    * Seeking input on regulations
    The U.S. Congress held hearings on AI between Sept. 11 and
13 and an AI forum featuring Meta Platforms  META.O  CEO Mark
Zuckerberg and Tesla CEO Elon Musk.
        More than 60 senators took part in the talks, during
which Musk called for a U.S. "referee" for AI. Lawmakers said
there was universal agreement about the need for government
regulation of the technology.
    On Sept. 12, the White House said Adobe  ADBE.O , IBM
 IBM.N , Nvidia  NVDA.O  and five other firms had signed
President Joe Biden's voluntary commitments governing AI, which
require steps such as watermarking AI-generated content.
    Washington D.C. district Judge Beryl Howell ruled on Aug. 21
that a work of art created by AI without any human input cannot
be copyrighted under U.S. law.
    The U.S. Federal Trade Commission (FTC) opened in July an
expansive investigation into OpenAI on claims that it has run
afoul of consumer protection laws.

 (Compiled by Alessandro Parodi and Amir Orusov in Gdansk; 
Editing by Kirsten Donovan, Mark Potter, Christina Fincher and
Milla Nissi)
 ((alessandro.parodi@thomsonreuters.com))