Mary-Ann Russon ​

Last night I was invited to go on the @BBCNews@flipboard.com channel with presenter Christian Fraser to react to the breaking news that #Meta & #Google had been found liable in a social media addiction trial in Los Angeles, brought by a young woman who claims she became addicted to social media as a child.

I had less than 15 minutes to get ready and I had to listen to the press conference and react to it immediately - so please excuse the messy hair 😅

Here are the main points of what I said:

📌This is a civil case. We need robust legislation, otherwise there is no precise US federal or state law that the tech giants have broken and they can keep appealing. You need to have a law in order to determine someone broke that law.

📌Right now, although we know there *is* definitely harm, the studies are inconclusive & only self-reporting. We need more studies and proven psychological analysis of harm to our young people.

📌We need our governments to stop being lenient with the tech giants and make it clear what specifically we expect from them, what tools and features we want to see, and we need to base that off specific, stated, measured harms.

#socialmedia #YouTube #Facebook #Instagram #techpolicy #technews #technology #socialmediaaddiction #socialmediaban

Very interesting new incident the global cybersecurity community is talking about right now, even Elon Musk has mentioned it... cybercriminals used AI-generated malware to poison a popular Python coding package called LiteLLM that helps manage AI API keys that provide access for more than 100 AI models to huge organisations including Nasa, Netflix, Nvidia, Adobe and Stripe.

The main reason the attack was discovered was that the malware was incredibly sloppily coded. If it hadn't been, no one would have noticed.

And the hilarious part was that the hackers got so upset that the cybersecurity researchers criticised the low quality of their code, even contacting one directly to argue that it was nevertheless indeed *successful* as millions of machines were infected 🤣

I have availability this week to provide commentary on this news and both the cybersecurity and AI aspects, so feel free to reach out😉

#cybersecurity #LiteLLM #AI #malware #cyberattack #Python #LLMs #technology #technews

I didn’t have going to New Broadcasting House this afternoon to be part of a 25-min @BBCNews@flipboard.com Verified Live programme chatting to Matthew Amroliwala about the UK government’s new plans to protect children on the internet on my 2026 bingo card, but it happened! 🎬

I’m reminded that it’s been 10 years (give or take a few months) since the lovely Mark Lappin escorted me to this same place for my first on-air appearance on the Tech Tent radio programme on BBC World Service, back when I was representing the International Business Times and snapped a shot of me similar to this one.

Feels like I’ve come full circle 😅
Video to come soon!

#technews #AI #generativeAI #socialmediaban #onlinesafetyact #UKlaw #technology

Phew, it's been a busy 24 hours!

A story this week from @theverge@flipboard.com in the US about a man accidentally being able to view thousands of DJI robot vacuum cleaners and access their data has been making waves over here across the pond.

I was on @BBCRadio4@social.bbc yesterday evening on the PM programme chatting to Evan Davis about the issue and the fact that everything (internet-enabled) continues to be hackable and you use it at your own risk.

Listen from 41:09 here 🎧🎙️:
https://bbc.co.uk/sounds/play/m002rqly

#robotvacuum #DJI #cybersecurity #technews #technology #IoT #InternetofThings

On Sunday I chatted to Geeta Guru-murthy live on @bbcworld@botsin.space about Clawdbot / Moltbot / OpenClaw & the website Moltbook, which is making the public think AI assistants are complaining about humans online 🤯

Not true! We don't need to fear an AI uprising just yet. Here's the full video🤖

Many thanks for allowing me to tech up the airwaves and really get into the ins and outs of the latest tech tea 🍵📺

#AI #Clawdbot #Moltbot #OpenClaw #Moltbook #Claude #Anthropic #AIassistants #technews #technology #analysis #LLM #technews #technology #cybersecurity #artificialintelligence

Pretty busy day but I've squeezed in a bit of time to summarise the the new US National Policy Framework for Artificial Intelligence that was released this afternoon by the White House.

While always dubious of stuff coming out of the Trump Administration, the document does look surprisingly sensible😅 They're definitely taking a page out of the discussions happening in Europe and UK in the last 2 months. Only a couple points I really dislike.

Here's a summary of key points (ordered by my preference of importance):

📌The Trump Administration is making "protecting children and adult victims from deepfake abuse" a key point. I was asked about these issues on the live BBC Your Voice panel discussion programme in mid-Feb.

📌Congress should create federal datasets and make them available for industry and academia in "AI-ready formats for use in training AI models and systems" [😬 Not sure about this! ]

📌Trump administration supports allowing AI models to be trained on copyrighted works, but they are happy for the Courts to work it out if you're not happy, so feel free to sue

📌Congress should consider working on "licensing frameworks or collective rights systems for rights holders to collectively negotiate compensation from AI providers, without incurring antitrust liability".

📌Nothing about preventing AI replacing your job, but American workers and youths should be given AI skills workforce training by Congress, and it should study trends in "task-level workforce realignment driven by AI"

📌Individual states are not allowed to penalise AI developers for unlawful conduct by a third party using their AI models [this raises a huge question about insurance🤔]

📌Small AI startups to be given grants and tax incentives to help boost AI deployment across the US

📌Congress should prevent the US government from compelling tech firms and AI providers to "ban, compel or alter content based on partisan or ideological agendas" and the public should be able to seek redress if they feel expression is censored on AI platforms

📌The US Congress is to insist on "robust" tools to help parents manage children's privacy settings on social media, as well as new features that reduce the risk of sexual exploitation and self-harm to minors

📌I absolutely adore this point:
"Congress should avoid setting ambiguous standards about permissible content, or open-ended liability, that could give rise to excessive litigation."

📌Congress should help existing law enforcement efforts to combat AI-enabled impersonation scams

📌Congress must ensure people don't pay increased electicity costs from new AI data centres being built, but should make federal permits easier to get for AI infrastructure construction

Paper: https://www.whitehouse.gov/wp-content/uploads/2026/03/03.20.26-National-Policy-Framework-for-Artificial-Intelligence-Legislative-Recommendations.pdf

#AI #generativeAI #techpolicy #WhiteHouse #Trump #scams #SocialMediaBan #copyright #copyrightinfringement #technews #technology

I've just finished reading the UK Government's new report today on the use of copyright works in the development of artificial intelligence (AI) systems, produced by the Dept for Science, Innovation and Technology, the Dept for Culture, Media and Sport & the Intellectual Property Office UK.

Here are the key respondent views + conclusions from the government - I'm available for on-air commentary as needed:

📌Gov UK most concerned about AI creating "digital replicas" of someone’s voice or face - but bizarrely not bothered about copyrighted texts, images, software etc being used in AI model training datasets (the report calls it "input transparency")

📌There is currently no mandate for AI developers to disclose publicly what IP they are using in their datasets, only which web crawlers they use to crawl the internet. But not what data they crawl or what it is used for

📌Creative industries strongly support the introduction of mandatory standards on input transparency, while tech firms have complained of the "practical and financial burdens of disclosure", particularly affecting small AI startups

📌AI developers & tech firms want exceptions to copyright for AI innovation. Creative industries have rejected the option to opt out of this

📌Gov UK to work on best practices around labelling the content outputted by AI models as AI-generated, but they want to see what other nations do first

📌However, Govt UK said this despite including in this report the new legislation from California, China and South Korea that already requires AI content to be labelled as such on release

📌Gov UK won't "intervene" in the copyright licensing market right now, because "many stakeholders" didn't want it (who tho? 👀)

📌Gov UK to "consider" further work on barriers to enforcing IP rights

Dunno, feels like the House of Lords is right. The UK Govt is still hedging against tightening legislation, in favour of the tech giants.

But what happens when the AI models ingest everything related to all white collar jobs that require a computer? What will we do then?

Full report: https://assets.publishing.service.gov.uk/media/69ba692226909a14239612e4/CP2602959_-_Report_on_Copyright_and_Artificial_Intelligence_web.pdf

#AI #copyrightlaw #UKlaw #generativeAI #technology #technews

Yesterday I attended a @Samsung@mastodon.cloud event in celebration of International Women’s Day and conversations about AI and women, where Samsung demonstrated how agentic AI is being used in their new S26 Ultra smartphones in new features.

These included further integration with Google Gemini for a more intuitive search engine to help reduce barriers to information, and NowNudge, an internal operating system feature that helps sum up WhatsApp messages, reminds users about meetings and helps the user to retrieve info and images on their device quickly.

I love learning about new tech and find it really interesting to attend events like this because you get to see not just what tech companies are doing, but also how they are explaining it to consumers and the words and framing they use.

And when there’s a particularly good product demonstrator or spokesperson like Patrick below, I love highlighting their explanations.

So in the interests of transparency and public learning, I invite you to be a fly on the wall and see from my POV in this quick video what I learned 😉

#AI #agenticAI #Samsung #SamsungS26Ultra #GoogleGemini #technology #technews

Following last night's appearance on @BBCRadio4@social.bbc, this morning I went on @bbcradioscot@bird.makeup's Breakfast programme to chat with Laura Maciver & James Cook about the @verge@mastodon.social story about a man accidentally being able to view thousands of DJI robot vacuum cleaners.

In particular, I mentioned that no matter what your business is, nowadays at the base level, every company in every industry is an IT company now.

Listen from 2:56:24 here 🎧🎙️:
https://bbc.co.uk/sounds/play/m002rnx7

#robotvacuum #DJI #cybersecurity #technews #technology #IoT #InternetofThings

Hooray! 🎉And now for some good news - Spain has followed in Australia's steps and is banning social media for under-16-year-olds starting from next week, and says it is due to meet with 4 other neighbouring nations considering doing the same to talk cross-border cooperative rules.

I was delighted to discuss this topic yesterday evening on @trtworld@bird.makeup's live Newshour programme with Rhiannon Jones. Really hopeful that this change sweeps across the EU and the UK, and then following that, new legislations are considered to protect children from being forced to become social media influencers too.

Watch the full video here 📺: https://www.youtube.com/watch?v=5YzHuEjHQGI

#socialmedia #socialmediaban #Spain #Australia #trtworld #meta #facebook #instagram #tiktok #youtube #technews #analysis #technology

This afternoon I bunked off ballet class to go on the @bbcworld@botsin.space channel and chat about Clawdbot / Moltbot and the fake Reddit-style website Moltbook that someone has made.

There are claims that AI bots are writing long posts complaining about humans and there are now posts going viral right now on Instagram, TikTok and YouTube claiming that they’re planning an AI uprising 🤖🤯

This is absolutely not true - the bots are simply replicating real Reddit posts they have been trained on.

I’ll post the full video tomorrow, happy to provide more commentary if anyone needs.

#Clawdbot #Moltbot #Moltbook #AI #AIassistant #technews #LLM #Anthropic #Claude #analysis #technology #tech

Popped onto BBC Radio Bristol this morning to chat to John Darvall about where the data on your smartphone goes and if it's safe, following the news that Lord Mandelson has been ordered to turn over his personal phone to the government.

Listen from 53:41 🎧🎙️:
https://www.bbc.co.uk/sounds/play/m002t04d

#cybersecurity #smartphone #iPhone #dataprivacy #iCloud #WhatsApp #technology #analysis #technews

Today the House of Lords has published an report on their inquiry into #AI and its impact on copyright and the creative industries, which lists some very sensible reasons why AI needs to be better regulated 🤖

This relates closely to what I said on @BBCNews@flipboard.com 2 weeks ago, that the House of Lords has consistently raised concerns about creative imagery being devalued and used by AI models for free in bills, yet somehow these bills disappear into nothing when they get to Parliament 🤔

Full report: https://publications.parliament.uk/pa/ld5901/ldselect/ldcomm/267/267.pdf

TL;DR in short, they found that:

📌 The answers produced by AI are not based on it learning anything, and thus shouldn’t be treated as such

📌 Gov UK needs to strengthen licensing, transparency & enforcement of copyright law, not weaken it

📌 Gov UK's mixed messaging regarding AI is creating big problems, including undermining investment

📌 Gov UK needs to issue a "clear public statement" that commercial AI developers need to obtain licences in order to use copyrighted works to train AI models

My analysis ✒️:

For some very strange reason, we don’t seem to value imagery, which makes no sense given copyright laws today.

The datasets used to train AI clearly need to be legislated. The reason that images were used to train AI models in the first place was to enable the AI to be able to differentiate different objects in a photo - i.e. to understand the difference between a car, a tree, a child, the grass in the background and the sky.

But AI models were never meant to replace creativity, art, design and illustration industries, the film industry, video games etc. All of these things are licensed and copyrighted. So why is nobody doing anything?

Why are we allowing computer algorithms to replace human expression when we have always enforced IP on human-created materials?

All governments should enforce legislation on the datasets being used to train the AI and I predict that firms in these industries and multiple other affected industries relating to text IP will eventually cobble together and sue these tech giants.

Because the alternative literally doesn’t make any sense.

#Copyright #HouseofLords #LLMs #technews #technology #UKlaw #analysis

Just been on @bbcworld@botsin.space news channel discussing with Kasia Madera about Instagram’s new rollout of a feature that alerts parents if their kids are searching for self-harm and suicide-related content (which, let’s be honest, they’re not doing that…)

If we’re being real, it’s much easier to find that kind of content on the internet, and while Meta definitely knows what users are searching for on their mobile device web browsers, it would be going down a slippery slope for us to permit further tracking and surveillance of children and young people, and eventually break GDPR etc.

So the only solution I can actually see would be to skip over that and just let the under-16s social media ban go through.

I’ll be on @BBC5Live@social.bbc in an hour to discuss the same thing.

#technology #technews #Instagram #privacy #socialmediaban #meta

On Monday, I was invited to film a 25-minute-long panel discussion episode of Your Voice with @BBCNews@flipboard.com live in the studio, answering questions viewers had sent in on the UK government's plans to update the Online Safety Act with further protections for children online, together with presenter Matthew Amroliwala, Ellen Roome MBE, Christopher Vallance and Jack Fenwick.

View the full video with everyone's contributions here:
https://www.youtube.com/watch?v=GivApfv13aY

#AI #generativeAI #LLM #UKlaw #OnlineSafetyAct #Grok #XAI #Claude #Gemini#ChatGPT #technology #technews

A couple of hours ago, Robert Peston released his podcast interview with the UK's AI minister Kanishka Narayan about today's launch of the UK Govt's first ever Sovereign AI fund 🤖

Here's a TL;DR news summary of what he said:

📌 The Sovereign AI website is an attempt at changing the "narrative" that the UK right now is "nowhere in AI" & only "focused on catching up and adoption, rather than being in the frontier".

📌 With AI, Govt UK wants to prevent a familiar story whereby entrepreneurs start great companies & then leave the UK when they think about scaling.

📌 It's not just about providing cash - it's also about visa support, procurement data + adapting public policy to help "anchor" more companies in the UK.

📌 Mr Narayan thinks income tax will become obsolete in 5-6 years, but the govt won't tax human labour, but it will "tax compute".

📌 UK govt sees the future of AI as "the single most important question for our economy right now" in terms of capital investment, jobs for the future, national security, drone defence systems & clean energy goals.

📌 UK doesn't have the capacity to take small companies global or "high risk early stage backers" of the scale seen in Silicon Valley. Govt capital needs to be provided earlier, in Round 2 or Series A of the seed funding process

📌 "I want this to sound and feel like a venture firm, which is the top tier firm." The fund will not be looking at primarily commercial applications of AI.

📌 He doesn't believe AI & robots will replace all people's jobs, even with the advancement of AI. Because people want to consume things, he feels that "wages of the things being consumed still in demand" will still go up, leading to jobs in those areas

📌He said the UK govt is "very rare" amongst global governments, in the fact it has access to AI models before they are deployed, to "test them for a series of risks", including the loss of control, where robots can run off and do things without human understanding or engagement

📌The fund is meant to make sure that AI models are "very much aligned with what we're trying to do as humans and as a society"

📌 The fund is partnering with the NHS, Accenture and Cognizant, large firms where the idea is to work together to design training to "quickly move the pace" of AI adoption in their firms

📌 He plans to launch technical excellence colleges focusing on digital tch and AI, because kids are already using AI, so a proper curriculum is needed

📌 However Mr Narayan doesn't believe in vibe coding, he feels it's still "really important" for software engineers to "understand fundamental architecture" so they can manage and oversee the coding agent

📌He's also soon to launch a campaign regarding chip design

📌 Money will be put into local authorities to accelerate planning for new data centres

Listen here: https://open.spotify.com/episode/5TpPYlVr9dh1JfPqAnhYYE

#AI #sovereignAI #UKlaw #techpolicy #technews #technology