🧵 New publication: How do people feel about regulating #AI in election campaigns? 🧵 In a new article, @adrauc.bsky.social, @kunkakom.bsky.social, and I examine when and why people support stronger AI regulation in political competition. www.sciencedirect.com/science/arti... 👇

Toured the new Oxford Humanities building today and it’s…humongous.

🗞️ 🤖 Weekend reading anyone? For the launch of @transformernews.ai as a standalone publication, they invited me to contribute a piece on what persuasive AI might mean for democracy and elections. Here’s the result… buff.ly/OJsNmpK

🗞️ 🤖 Weekend reading anyone? For the launch of @transformernews.ai as a standalone publication, they invited me to contribute a piece on what persuasive AI might mean for democracy and elections. Here’s the result… buff.ly/OJsNmpK

Had a great time soft-launching our forthcoming research report on generative AI and news (out October 7th) at the b° future festival in Germany's former capital, Bonn – wonderfully organised by the Bonn Institute team around Ellen Heinrich. Auf Wiedersehen

Another one of those weeks in which the same news media engaged in lots of (often justified) handwringing about errors in AI outputs abysmally fail in properly framing the misleading or plainly false statements by various political elites and leaders.

cho·ti·nerd | ˈtʃoʊ.ti.nɜrd | verb (transitive) To systematically and subtly dismantle someone’s arguments or credibility during an interview with such surgical precision that the subject remains unaware of how they have completely ridiculed themselves until after the fact.

Freedom of speech is now apparently woke in the US and needs to be cancelled, am I getting this right?

Always a bit awkward to do these posts but I was delighted to learn at the Future of Journalism Conference last week that I won @digitaljournalism.bsky.social's Outstanding Article of the Year Award for my paper on how AI reshapes the news' dependency on platforms companies.

Interesting to see this against audience views on the topic. Scepticism of AI use in news remains, but growing acceptance of AI assisted news-creation (which this might fall under) – see @reutersinstitute.bsky.social data from this year and last year on comfort around such uses.

Feels like as good a time as any to get out the old Frank Wilhoit adage that US conservatism today basically amounts to the idea that “There must be in-groups whom the law protects but does not bind, alongside out-groups whom the law binds but does not protect”

Finally found where they make them

Using the big technology of the 19th century to get to Cardiff to talk about the big technology of the 21st (Train ride is bumpy, somehow there’s a metaphor in this) — looking forward to the Future of Journalism conference and seeing colleagues old and new.

The birth of a new edition: Exclusive behind the scenes footage of how the @reutersinstitute.bsky.social Digital News Report gets made.

Perhaps the final meeting from the old @oii.ox.ac.uk buildings before the institute moves to its new haunt at the Schwarzman Centre for the Humanities – featuring the great Chico Kerche, @rocher.lc, Franziska Hafner and @lujain.bsky.social to talk about information and news within LLMs.

To my French friends, a quick question: Is AI Overviews available in France? If not, why?

Think a lot of people assume that if the AI bubble bursts, the things they hate will go away, and I worry that’s almost exactly the wrong way around

I have become genuinely fascinated by the work of @birchlse.bsky.social & colleagues and it has certainly shifted my own priors somewhat…looking forward to reading this.

How will I disillusion myself tonight?

New paper with @kirstenaeddy.bsky.social in Journalism on whether people across 45 countries think news coverage of people their age, their gender, and with their political views is fair or unfair. We outline some interesting demographic differences (more details in 🧵)... doi.org/10.1177/1464...

NEW - Media Platforming and the Normalisation of Extreme Right Views - cup.org/4mmVIAL "exposure to uncritical interviews increases agreement with extreme statements and perceptions of broader support in the population" - @dianebolet.bsky.social & @florianfoos.bsky.social #OpenAccess

Was reminded of Bentham’s argument that fear of crime can be a worse harm than crime itself. Everyone can be affected by fear of crime but relatively few people experience serious crime. And fear of crime can easily be manipulated for nefarious purposes.

I only met Dame Stephanie “Steve” Shirley twice and both times she was talking about her life story which is among the most impressive I know. The world is poorer without her. www.bbc.co.uk/news/article...

Just leaving this here www.un.org/unispal/docu... ihl-databases.icrc.org/en/customary...

Quick reminder from Semafor's Reed Albergotti that OpenAI's release of open-weight (not open source) models is still about business: "By offering its own models, it can both keep customers inside its family of products, and benefit from improvements and new ideas shared by users."

"Growing big quickly is no guarantee of success, in fact, it's often a liability. Starting up as a behemoth takes the focus away from audiences and their information needs and ultimately, undermines financial sustainability." – always worth reading Peter Erdelyi. buff.ly/hBwSpFg

Again and again comes down to two questions: 1) Is the EU willing to put their money where their mouth is and 2) what are they willing to sacrifice in a world where you can’t fund/do everything

Ah Oxford, never change

Great piece from @himself.bsky.social about his new paper with @hahrie.bsky.social for our AI & Democratic Freedoms series - link below. www.programmablemutter.com/p/the-ai-dem...

So when are we starting to monitor US GDP with nightlights?

"The general message from the tech giants is that this doesn’t matter: they have the audience, they’re creating the infrastructure, they’ll eventually figure it out. Moving fast, however, will be critical." buff.ly/c6qC6kY

Having now watched t̶h̶e̶ m̶o̶n̶s̶t̶r̶o̶s̶i̶t̶y̶ t̶h̶a̶t̶ i̶s̶ "My Oxford Year", I've devised this little bullshit bingo that will reliably knock your brains out by minute 12 if you choose to combine it with a drinking game

Really quite embarrassing how bad Siri is at the easiest tasks. Nailing easy calendar entries (see screenshot) should not even be a bar but the floor and even that it cannot really do.

Still two months away, but I will be at the @bonninstitute.bsky.social festival in Bonn on 3rd October, presenting forthcoming @reutersinstitute.bsky.social research (with @richardfletcher.bsky.social & @rasmuskleis.bsky.social) on what audiences think of genAI (in news). Say hi! buff.ly/z2tw1WD

YouGov has some interesting new data on how people in the UK see AI. AI use is fairly common now but we still have 38% say they never (consciously) use AI for leisure, work, or study (this will differ from actual use but still).

If the NYT's review of Oxford's new hotel in an ex-public toilet only mentioned that its a) in the middle of a busy road, b) dangerous & inaccessible to get to & c) doesn't really make an effort to flag this so people end up confused at the OII next door. buff.ly/ABbU6mn

Listing this as the world's biggest risk (ahead of pollution, climate change, large scale wars, bio diversity collapse) seems like the type of thing that those who exacerbate these other risks would want us to focus on instead...

*silent internal screaming*

Hulk Hogan dead? The White House is really pulling out all the stops to make the Epstein thing go away.

Alarmist discourse about the effect of generative AI on elections risks distorting policy, damaging trust, and distracting from the real issues. Read the latest @oii.ox.ac.uk blog post from @felixsimon.bsky.social and @sachaltay.bsky.social

Worth noting this describes the credit card route but not the face scan route, which is via a different provider. (And IME is quicker)

What are the risks of a discourse that exaggerates the role and risk of GenAI around elections? We look at this question in more detail in another excerpt from @sachaltay.bsky.social & my paper on GenAI and elections, this time kindly published by the @oii.ox.ac.uk.

We keep seeing these articles suggesting privacy is dead and we can all be tracked by all sort of apparatuses. We did the maths earlier this year in an article in Nature Communications, and none of this holds.

Genuinely one of the best social media innovations to come out of a mainstream newsroom in ages – showing you can package content different and capture new audiences, and even have some fun with it. Real loss for the Post, but genuinely interested to see what Dave and co do next…

Stark warning from the AFP journalists association "Since AFP was founded in August 1944, we have lost journalists in conflicts, we have had wounded and prisoners in our ranks, but none of us can recall seeing a colleague die of hunger" www.aa.com.tr/en/middle-ea...

Just from a quick look, this new study from Kobi & colleagues is mindblowing. - Conversational LLMs are persuasive & they are so because they are good at using facts and evidence that support their arguments - the information used to persuade was mostly accurate

Ben’s work has been incredibly helpful for @sachaltay.bsky.social and me in writing about GenAI and elections, so this this praise coming from him means a lot and you should read his papers!

Repeat after me: Elites who are careless or actively hostile towards the press and democracy are a bigger issue for the latter than AI.