AI Weekly Digest - December 23 2024

UK creatives/media reject plan for AI groups' copyright exemption

Representatives of the UK creative industries and the media have rejected the government’s plan to create a copyright exemption to help AI groups train their algorithms. Newspapers, publishers, writers, musicians, photographers and film producers issued a joint statement on behalf of thousands of workers. This opposed a proposal by ministers to allow the likes of OpenAI, Google and Meta to train their AI systems on published works - unless their owners actively opt-out. The Creative Rights in AI Coalition (Crac) said existing copyright laws must be respected and enforced, rather than degraded. Those opposing the government plan include the Guardian, Financial Times, Telegraph, Getty Images, the Daily Mail Group and Newsquest. Other members are British Phonographic Industry, the Independent Society of Musicians, the Motion Picture Association, the Society of Authors and Mumsnet.

Government consultation on use of copyrighted material to train AI models

The creatives' pushback came after the UK government launched a 10-week consultation on the use of copyrighted material to train artificial intelligence models. Culture secretary Lisa Nandy said: "This government firmly believes that our musicians, writers, artists and other creatives should have the ability to know and control how their content is used by AI firms and be able to seek licensing deals and fair payment. Achieving this, and ensuring legal certainty, will help our creative and AI sectors grow and innovate together in partnership."

Perplexity triples valuation to $9bn after fourth 2024 funding round

AI search start-up Perplexity's valuation has tripled to $9bn as it raised $500m in its fourth funding round this year. Sources told the FT that the group had been inundated with unsolicited interest from new investors keen to secure exposure to AI.

New Google EMEA chief Weinstein's AI focus

Google has named Debbie Weinstein as its president in Europe, the Middle East and Africa. Weinstein, who had been the US tech giant's vice-president and MD in the UK and Ireland, said her focus will be on “unlocking AI-powered growth for everyone”. Weinstein said: “Europe, the Middle East and Africa is an amazingly diverse and varied region, but the enormous growth opportunity that AI can create is universal." She added: “I’m excited to be stepping into this role at a pivotal time, in a company where I’ve spent the last 10 years and leading a region where I’ve spent much of my life.”

Google business chief hails 'pivotal moment for technology'

Google senior vice-president and chief business officer Philipp Schindler said of Debbie Weinstein's appointment: “This is the AI era and we are only just beginning to see its transformative impact on business and society. In such a pivotal moment for technology, I’m thrilled we’ve appointed a visionary leader to be our president of Google EMEA. Debbie brings a track record of unlocking growth that benefits everyone, alongside the passion and focus needed to help our customers succeed, as we bring the best of Google’s Gemini-era to everyone across EMEA.”

BBC World Service 'grappling' with the issue of AI

BBC News global director Jonathan Munro told the Culture, Media and Sport Committee that the World Service is “grappling” with the issue of AI. Munro said: “We currently have 42 language services. It’s perfectly possible to see that growing significantly over time, with technology allowing us to spread our journalism more prolifically around parts of the world where that is not readily available apart from in English. We are looking at that proactively, but I don’t think it’s a big leap moment. We want to tread relatively carefully because the reputation of the BBC is sacrosanct and want to make sure when we move into that field, we learn lessons and adapt models from market to market.”

YouTube tests ways for celebrities to find AI-generated content

Google-owned YouTube is partnering with the Creative Artists Agency to test ways to allow celebrities and athletes to find AI-generated content on the video sharing platform using their likeness. They will then be able to submit requests for the content to be removed. The company said the tests will begin early next year before being rolled out to “top YouTube creators, creative professionals, and other leading partners representing talent".

BBC complains over Apple Intelligence generating false headline

Apple declined to comment after the BBC lodged a complaint about Apple Intelligence generating a false headline about a US murder. The iPhone AI feature's summary suggested BBC News published an article wrongly stating that Luigi Mangione, who was arrested following the murder of healthcare insurance CEO Brian Thompson, had shot himself. A BBC spokesperson said it had contacted Apple "to raise this concern and fix the problem".

Writers Guild demands 'immediate legal action' from film/TV groups

The Writers Guild of America has demanded film and TV groups take “immediate legal action” to prevent AI firms training their models on members' work. In a letter to the CEOs of entertainment groups including Warner Bros. Discovery, Disney and Paramount Global, the Guild said: “It’s time for the studios to come off the sidelines. After this industry has spent decades fighting piracy, it cannot stand idly by while tech companies steal full libraries of content for their own financial gain.”

Elon Musk's xAI 'sells stake to Oman'

Elon Musk’s xAI is reported to have sold a stake to Oman. The country’s state news agency said the Oman Investment Authority has acquired a stake in the artificial intelligence company. The sovereign wealth fund also holds shares in Musk's SpaceX.

Meta: 'OpenAI’s conduct could have seismic implications for Silicon Valley'

Meta has urged California’s attorney general to block OpenAI’s planned for-profit switch, saying clearance would set a dangerous precedent. It said: “OpenAI’s conduct could have seismic implications for Silicon Valley. If OpenAI’s new business model is valid, non-profit investors would get the same for-profit upside as those who invest the conventional way in for-profit companies while also benefiting from tax write-offs bestowed by the government.”

OpenAI: Musk backed for-profit structure as early as 2017

OpenAI has released an email showing then-co chairman Elon Musk backed the group adopting a for-profit structure as early as 2017. Musk has taken legal action this year to prevent the ChatGPT developer going for-profit, and the group said in a statement: “When he didn’t get majority equity and full control, he walked away and told us we would fail. Now that OpenAI is the leading AI research lab and Elon runs a competing AI company, he’s asking the court to stop us from effectively pursuing our mission."

Arm CEO: AI usage on smartphones 'to happen gradually and then suddenly'

Arm CEO Rene Haas said he expects AI usage on smartphones is "something that is going to happen gradually and then suddenly", forecasting a "knee-in-the-curve moment" when chips are capable of handling new apps. He said: "It takes two or three years to develop a chip. Think about the chips that are in that new iPhone when they were conceived, when they were designed, and when the features that we thought about had to go inside that phone. ChatGPT didn’t even exist at that time."

Penetration rate for AI-enabled PCs set to soar

Morgan Stanley forecast the penetration rate for AI-enabled PCs will rise from 1%-2% this year to 65% in 2028, boosted by unlimited personalisation, and greater privacy and security.

OpenAI researcher-turned-whistleblower found dead

OpenAI researcher-turned-whistleblower Suchir Balaji has been found dead in his San Francisco apartment. Balaji, who alleged the ChatGPT-maker violated US copyright law while developing the chatbot, told the Wall Street Journal in October that "technologies like ChatGPT were damaging the internet". The San Francisco medical examiner's office determined his death to be suicide with police finding no evidence of foul play. OpenAI said it was "devastated to learn of this incredibly sad news today and our hearts go out to Suchir's loved ones during this difficult time".

Altman donates £1m to Trump’s inaugural fund

OpenAI chief executive Sam Altman has become the latest US tech tycoon to pledge $1m to Donald Trump’s inaugural fund, with AI search start-up Perplexity also committing $1m ahead of next month's inauguration.

Groups join Creative Rights in AI Coalition

About 40 industry groups, including the News Media Association, the Society of Editors and PRS for Music, have joined new umbrella body The Creative Rights in AI Coalition, calling for government action to defend copyright. The body said: "The UK creative industries generate well over £100bn annually. We have, quite literally, earned the right to have our voice heard. The key to that success, and future growth, is copyright law.”

UKAI calls on government to create dedicated regulator

New artificial intelligence trade association UKAI has called on the government to create a dedicated regulator for the burgeoning technology. The industry body said AI needs its own regulator as it will become a substantial part of the economy, requiring a level of oversight similar to the Financial Conduct Authority’s regulation of financial services. A UKAI spokesperson said: “An AI regulator would have a clear focus to interpret existing regulation, as well as co-ordinate AI regulation across sectors, supporting existing regulators and providing unified guidance, avoiding duplication and inefficiency. Without an AI-specific regulator we risk a lack of clarity and the likelihood that regulation will be disjointed. Existing regulators have finite resources and lack the extra capacity that will be required to plan for and oversee new AI regulation.”

Instagram denies knowingly hosting accounts advertising abuse material

Meta said "child exploitation of any kind is horrific and we have clear rules against it", after charity the 5Rights Foundation accused Instagram of knowingly hosting accounts which advertise AI-generated child sexual abuse material. The charity told Ofcom and the Information Commissioner's Office: "Meta is failing to take meaningful steps to suitably address the spread of illegal content. The scale of victimisation... is vast."

ChatGPT Search available to all users

OpenAI has made ChatGPT Search available to all users, after previously restricting it to paid tiers. Users will also be able to embed it as the default search engine within the web browser of their choice.

Meta Ray-Bans enable real-time language translation

Meta has added AI video capability and real-time language translation functionality in an update to its Ray-Ban smart glasses. They will be able to translate speech between English and Spanish, French or Italian.

Sutskever: Advent of reasoning capabilities will make AI technology far less predictable

OpenAI's former chief scientist has warned that the advent of reasoning capabilities will make AI technology far less predictable. Ilya Sutskever said: "Pre-training as we know it will unquestionably end. While compute is growing, the data is not growing, because we have but one internet." Sutskever, who co-founded Safe Superintelligence, was a key player in Sam Altman's short-lived ousting from OpenAI. He told Canada's NeurIPS conference: "The more it reasons, the more unpredictable it becomes. The chess AIs, the really good ones, are unpredictable to the best human chess players."