Friday, May 26, 2023

OpenAI boss downplays fears ChatGPT maker could leave Europe over AI rules

KELVIN CHAN, 
AP Business Writer
May 26, 2023
2of8'OpenAI's CEO Sam Altman, the founder of ChatGPT and creator of OpenAI gestures while speaking at University College London, as part of his world tour of speaking engagements in London, Wednesday, May 24, 2023.
Alastair Grant/APShow More


LONDON (AP) — OpenAI CEO Sam Altman on Friday downplayed worries that the ChatGPT maker could exit the European Union if it can't comply with the bloc's strict new artificial intelligence rules, coming after a top official rebuked him for comments raising such a possibility.

Altman is traveling through Europe as part of a world tour to meet with officials and promote his AI company, which has unleashed a frenzy around the globe.

At a stop this week in London, he said OpenAI might leave if the artificial intelligence rules that the EU is drawing up are too tough. That triggered a pointed reply on social media from European Commissioner Thierry Breton, accusing the company of blackmail.

Breton, who's in charge of digital policy, linked to a Financial Times article quoting Altman saying that OpenAI “will try to comply, but if we can’t comply we will cease operating.”

Altman sought to calm the waters a day later, tweeting: “very productive week of conversations in europe about how to best regulate AI! we are excited to continue to operate here and of course have no plans to leave.”

The European Union is at the forefront of global efforts to draw up guardrails for artificial intelligence, with its AI Act in the final stages after years of work. The rapid rise of general purpose AI chatbots like ChatGPT caught EU officials off guard, and they scrambled to add provisions covering so-called generative AI systems, which can produce convincingly human-like conversational answers, essays, images and more in response to questions from users.

“There is no point in attempting blackmail — claiming that by crafting a clear framework, Europe is holding up the rollout of generative #AI,” Breton said in his tweet. He added that the EU aims to “assist companies in their preparation” for the AI Act.

Altman tweeted that his European tour includes Warsaw, Poland; Munich, Germany; Paris; Madrid; Lisbon, Portugal; and London. Brussels, headquarters of the EU, has not been mentioned.

He has met with world leaders including British Prime Minister Rishi Sunak, French President Emmanuel Macron, Spanish Prime Minister Pedro Sanchez and German Chancellor Olaf Scholz.

Google CEO Sundar Pichai also has been crisscrossing Europe this week to discuss AI with officials like Scholz, European commissioners including Breton, Swedish Prime Minister Ulf Kristersson, and two EU lawmakers who spearheaded the Parliament's work on the AI rules.

“Good to discuss the need for responsible regulation and transatlantic convergence on AI,” Pichai tweeted.

Google has released its own conversational chatbot, Bard, to compete with ChatGPT.

Other tech company bosses have been wading into the debate this week over whether and how to regulate artificial intelligence, including Microsoft President Brad Smith, who unveiled a blueprint for public governance of AI on Thursday.

Microsoft has invested billions in OpenAI and integrated ChatGPT-like technology into its products, including a chatbot for its Bing search engine.

Altman told congressional lawmakers this month that AI should be regulated by a U.S. or global agency because increasingly powerful systems will need government intervention to reduce their risks.

Altman was mobbed by students when he appeared in a “fireside chat” at University College London on Wednesday. He told the audience that the “right answer” to regulating AI is “probably something between the traditional European, U.K. approach and the traditional U.S. approach.”

“I think you really don’t want to overregulate this before you know what shape the technology is going to be," Altman said.

There's still potential to come up with “some sort of global set of norms and enforcement," he said, adding that AI regulation has been a “recurring topic” on his world tour, which has also included stops in Toronto, Rio de Janeiro and Lagos, Nigeria.

OpenAI CEO rolls back threat to quit Europe over regulation


Sam Altman sparked criticism after accusing the European Union of "overregulating" artificial intelligence platforms. Having warned that ChatGPT may pull out of Europe, Altman now says his firm has "no plans to leave."


OpenAI — the company behind the popular artificial intelligence (AI) platform ChatGPT — has no plans to leave Europe, CEO Sam Altman said on Friday, reversing a threat made earlier this week.

"We are excited to continue to operate here and of course have no plans to leave," Altman said in a tweet.



The remark is a U-turn from comments made to journalists on Wednesday where he accused the European Union's proposed AI Act of potentially "overregulating" platforms like his.

"The current draft of the EU AI Act would be overregulating, but we have heard it's going to get pulled back," Altman told Reuters. "They are still talking about it."

Altman was cited by the Financial Times as saying that the draft EU rules were causing him "a lot of concern" but that OpenAI would indeed try to comply with them. "But if we can't comply with them, we will cease operations [in Europe]."

During an event at the Technical University of Munich (TUM) on Wednesday, Altman said that he thinks typically, regulators should take a wait-and-see approach, but noted how AI regulation is an area "where we should be proactive."


Brussels says AI rules 'cannot be bargained'


Altman's initial threat was met with a strong response from Brussels, with European Internal Market Commissioner Thierry Breton insisting that Europe's AI rules are not up for negotiation.

Breton said there was no point making speculative threats and claiming Europe was holding up the rollout of generative AI, referring to the type of AI that can create text, images and other content.

"Let's be clear, our rules are put in place for the security and well-being of our citizens and this cannot be bargained," Breton told Reuters.


"Europe has been ahead of the curve designing a solid and balanced regulatory framework for AI which tackles risks related to fundamental rights or safety, but also enables innovation for Europe to become a frontrunner in trustworthy AI," he said.

Microsoft's President Brad Smith said during an event in the United States on Thursday that he was optimistic "reason will prevail" and that the final AI Act would be an acceptable compromise.

What is ChatGPT and why is it so controversial?


ChatGPT is an AI platform that is able to write emails, essays and software code with minimal human guidance. It can also give detailed answers on a wide range of topics.

Microsoft has invested nearly $10 billion in Open AI, the laboratory behind ChatGPT, while Google and Chinese tech giant Alibaba have rushed out their own versions.

The arrival of such chatbots has surprised many policymakers and business leaders, who hadn't expected AI technology to advance so quickly.

Concerns are now rising that AI may grow too powerful, cause hundreds of millions of job losses and be used for nefarious purposes.

Critics say it will decimate entire industries, lead to a flood of misinformation and copyright infringements, and entrench race, gender and political biases.

AI's supporters claim that the technology will improve lives by doing menial tasks better and revolutionize human interaction with machines.

When will the European AI Act be passed?


The EU is in the final stages of approval of its regulations to cover generative AI tools, such as OpenAI's Chat GPT.

After being sent back to the drawing board by the arrival of ChatGPT, EU lawmakers reached common ground on the draft of the act earlier this month.

EU lawmakers and EU national governments are expected to thrash out the details of the rules in the coming months before they become legislation that may become the global standard for the technology.

The final law isn't expected to take effect until 2025.

mm/sri (dpa, Reuters)

No comments:

Post a Comment