DW Innovation

Ancient Egypt or Pompeii – more and more #immersive exhibitions with a focus on #history are popping up.

But do people actually enjoy them?

Find out what visitors had to say in this short German-language video report by Jochen Lohmann for DW:

👇 […]

Case Studies — JournalismAI

There are so many #papers on #AI and #journalism by now, it's really hard to keep track of them.

Conveniently, the JournalismAI project has set up a #database to collect and tag the most interesting studies:

👇
https://www.journalismai.info/resources/case-studies

"This is a textbook case of what happens when there's this conflict between what companies want, what nonprofit individuals want, how much responsibility people have when they take money, who gets control and when. How much democracy versus just 'I have the power to do something, therefore I'm […]

It's true: AI tools can offer helpful starting points for visual investigations – but also cause a lot of confusion.

A deep dive by Aisvarya Chandrasekar and Klaudia Jaźwińska for the Columbia Journalism Review:

https://www.cjr.org/tow_center/why-ai-models-are-bad-at-verifying-photos.php […]

Food insecurity: Nutrition and geography in a hungry world – DW – 07/28/2025 More than 2.3 billion people experience moderate or severe food insecurity, and 673.2 million are undernourished, according to a global report by UN agencies. DW examines hunger and unequal food access over the years

#DDJ #DataViz

4 charts and 1 map that explain the global problem of unhealthy diets, food insecurity, and sheer hunger.

(via DW Data)

https://www.dw.com/en/food-insecurity-nutrition-and-geography-in-a-hungry-world/a-73435007

"Imagine a world where scientific experiments yielded different results every time you ran them, even with identical setups. (...) In most fields, such inconsistencies would be unacceptable. Yet, in the (.) world of large language models (LLMs), we’ve often accepted a curious paradox: the same […]

The #CfP for the 2026 Languages & The #Media #Conference just opened.

The event will be hosted in London, and the main theme will be:

"Moving Images That Move Audiences – Localizing with Intent"

The deadline for proposals is January 16th, 2026.

More info here […]

Content authenticity and digital markers vs. #disinfo: The BBC’s R&D team has been testing Sony’s new #C2PA #video camera.

Results presented by Judy Parnall and Charlie Halford.

👇
https://www.bbc.co.uk/rd/articles/2025-09-news-content-verification-credentials-trust

C2PA will also play a role […]

"This dance of refusal and counter-refusal, called #taarof, governs countless daily interactions in #Persian culture. And #AI models are terrible at it."

(via Benji Edwards and @arstechnica)

https://arstechnica.com/ai/2025/09/when-no-means-yes-why-ai-chatbots-cant-process-persian-social-etiquette/

"... you might recall the feeling of confusion after opening such a document, followed by frustration – Wait, what is this exactly? – before you begin to wonder if the sender simply used AI to generate large blocks of text instead of thinking it through. If this sounds familiar, you have been […]

5 Best VPN Services (2025), Tested and Reviewed Every VPN says it’s the best, but only some of them are telling the truth.

Virtual private networks have become standard tools for many journalists, as the services promise to enhance privacy, security, and freedom.

Jacob Roach tried to find a #VPN that really works.

(via @WIRED)

https://www.wired.com/gallery/best-vpn/

Why is #synthetic #audio still so hard to detect? 🎧

Because there are SO many ways to create it!

But fear not. Our colleagues Julia Bayer and Anna Schild have set out to test and compare three freely available tools: Hiya, DeepfakeTotal and DeepFake-O-Meter.

👇
Check out our latest blogpost […]

Map of the Internet in 1977 with white dots (IMPS), rectangles (host computers). and jagged lines (connections to other networks). Credit: The Computer History Museum.

If you're into #longreads and tech #history, check out this 3-part series on the people and ideas that made the #Internet by @arstechnica:

I: An ARPA Dream Takes Form […]

[Original post on mastodon.social]

Cat soap operas and babies trapped in space: the ‘AI slop’ taking over YouTube Nearly one in 10 of the fastest growing channels globally consist of mass-produced, surreal AI-generated videos

Just in case you missed it: Priya Bharadia's sobering piece on #YouTube #AIslop for @guardian.

👇
https://www.theguardian.com/technology/2025/aug/11/cat-soap-operas-and-babies-trapped-in-space-the-ai-slop-taking-over-youtube

From Walled Gardens to World Grid: A Short History of How We Centralized the Internet | HackerNoon From ARPANET’s open roots to today’s walled gardens, explore how we centralized the internet—and why the fight for a freer web isn’t over.

The internet was born "free" – but today it’s dominated by corporate "walled gardens."

Can we reclaim a global, open grid?

"Ibukun OG" for @hackernoon.

👇
https://hackernoon.com/from-walled-gardens-to-world-grid-a-short-history-of-how-we-centralized-the-internet

#internet #history

Turning 3D models into fully interactive experiences – with hotspots and everything?

Our colleagues at XReco have developed a tool that makes this possible: The 3D infographic template.

Clickable, zoomable & ready to insert into your next online article.

Check out their blogpost to find out […]

Irreparable #SocialMedia?

"In social science, we always teach that everything is a normal distribution.

The move from studying the conventional social world to studying the online social world means that you're moving from these nice normal distributions to these horrible power law […]

"Today’s generative platforms have learned a lot from the sheer volume of information they have gobbled up, but tomorrow’s agentic systems are going to need common sense specific to an individual person’s context, which can’t be learned from large data sets."

Thoughts on the next #AI […]

#ChatEurope is a web-based #news aggregator with an LLM-driven chatbot wired into it.

The #AI tool under-delivered at first, but has gotten a lot better – thanks to a committed consortium that wants to build a truly useful and trustworthy service […]

Grin and pair it: Our Smile Tracker loves to find the funny BBC R&D; has created the Smile Tracker tool to help showcase the great comedy on BBC iPlayer.

This one is about BBC comedies. And R&D. And marketing. And dating.

Robert Dawes introduces the "Smile Tracker". 😄

https://www.bbc.co.uk/rd/articles/2025-05-smile-tracker/

"The (human) host will get less smiley, less engaging, and so on. The virtual human is very standardized in terms of attitude."

@jasonkoebler on the advancement of salesperson avatars in China […]

Generative AI models love to cite Reuters and Axios, study finds Muck Rack found that major models cite journalism in nearly half of responses that require "recency."

Claude, Gemini, and ChatGPT refer to journalistic sources more often than one would think – and they prefer specific media outlets.

(via @niemanlab)

https://www.niemanlab.org/2025/07/generative-ai-models-love-to-cite-reuters-and-axios-study-finds/

#AI #journalism #news

How AI Affects Our Minds: Early Evidence From Harvard and MIT | HackerNoon Discover how LLMs may impact memory, motivation, and critical thinking. MIT and Harvard studies reveal surprising cognitive effects.

Could you cut it in the "brain only" group?

Vik Bogdanov has checked recent #Harvard and #MIT research – and explains how #AI tools can (negatively) affect our minds.

https://hackernoon.com/how-ai-affects-our-minds-early-evidence-from-harvard-and-mit

(via @hackernoon)

#neuroscience

From watermarks to missing sources and credibility gaps, here's how you can sharpen your senses – and avoid falling for the latest wave of newscast fakes:

https://www.dw.com/en/fact-check-how-to-spot-ai-generated-newscasts/a-73053782

Useful media and information literacy feature by Monir […]

"Editor's note: This essay has been removed." 🫣

(via Press Gazette)

https://pressgazette.co.uk/news/ai-hoaxer-a-fake-x-account-and-the-case-of-margaux-blanchard/

#fake #hoax #AI

NYC’s Urban Textscape Analyzing All of the Words Found on NYC Streets

"What if you could search every visible word on New York City’s streets?

(...)

The data is astonishing; it feels like sifting through the city's source code."

Impressive mix of #DDJ and #scrollytelling by @thepudding

https://pudding.cool/2025/07/street-view/

Trusted news sites may benefit in an internet full of AI-generated fakes, a new study finds An economics paper found subscriber retention and daily visits both increased after readers were confronted with a difficult quiz with AI-generated images.

"If something — let’s say trustworthiness — becomes really scarce, then it becomes very valuable."

(via @niemanlab)

https://www.niemanlab.org/2025/08/trusted-news-sites-may-benefit-in-an-internet-full-of-ai-generated-fakes-a-new-study-finds/

#journalism #trust

Will your next trip be virtual?

Using different technologies, our colleagues at XReco turn buildings into 3D models – and put them in virtual studios.

Whether you want to place yourself inside a landmark or make it rotate on a platform in front of you, XReco aims to facilitate the process of […]

Just a quick announcement that our colleagues Eva Lopez and Anna Schild will be the hosts of an upcoming
#EBU #webinar. It's called:

"Co-creating veraAI #FactChecking #tools for and with media professionals"

The session is scheduled for:
03.09.25, 11:00h (Geneva time).

More info and […]

Dear #Verification, #GeoLocation and #OSINT people:

We've built a new #tool! 🙌

It's called <You Stitch It> and allows you to...

...create panoramas from videos and photos... 🌆
...magnify single frames... 🖼️
...and download them. 💾

It's free to use, of course. And you're more than welcome to […]