Thank you for having me, @fbk-mt.bsky.social! It was great fun to share some of my current work and thinking, and I'm very grateful for your feedback, thoughtful questions and timely suggestions!

Love this piece on language equity in content moderation tools. The authors argue that rather than focusing on investment equity across languages, como resources and interventions should be motivated by RISKS, e.g. social vulnerability and degree of reliance on the system.

This was one of my favourite papers from EMNLP last October - the authors curate over 14,000 counterfactual sentences in English and Spanish for investigating diverse and often intersectional social biases in LLMS across these ten social dimensions ⤵️

Really love this comparison of lack of AI accountability today to unfair credit reporting practices in the 1960s. Andrews argues that the FRCA effectively built an "epistemic layer" to protect people by defining terms, imposing accuracy standards, and creating an audit trail...

Do multilingual large language models (LLMs) associate particular traits with men and women? How can we measure these gendered stereotypes across multiple languages? In our latest work, we investigate 16 gendered stereotypes in LLMs in 30 European languages 🧵

Do LLMs encode gendered stereotypes? Do they encode the same stereotypes across languages? How can we better measure gender bias in multilingual LLMs? I'll be presenting a poster on my work on these questions at this research showcase tomorrow hosted by @uoe-gender-ed.bsky.social 👇 See you there!

Had an amazing time at #NAACL last week in Albuquerque 🌵 Especially enjoyed presenting my recent paper on machine translation for Guinea-Bissau Creole at the workshop on MT for low-resource languages aclanthology.org/2025.loresmt... 🎉 thanks to organisers, reviewers & all who gave feedback! 🙏

Enjoying this excellent writing from Jason Edward Lewis, Hēmi Whaanga and Ceyda Yolgörmez. Paper: Abundant intelligences: placing AI within Indigenous knowledge frameworks, AI & Society, 2024, link.springer.com/article/10.1...

Language policies have started wars and led to the deaths of thousands of people worldwide. This EO will mark the first time the US will have a national language and will restrict *millions* from accessing healthcare, benefits, and information: abcnews.go.com/Politics/tru...

If participatory approaches are going to impact the future trajectory of AI, we need to invest in building the field, write @timdavies.org.uk and @jenitennison.com.

Thoughtful examination here of the risks of touting "low-resource NLP" as a good in its own right, without engaging with and empowering language communities

Amazing to see the first language and culture textbook in Cape Verdian creole (Kabuverdianu) being rolled out for 10th graders, building their metalinguistic awareness and historic and scientific knowledge about the language: 📣 inforpress.cv/pt/governo-a... 📖 minedu.gov.cv/manuais62

Great overview by @chrischirp.bsky.social categorising 76 Trump administration actions from the last 3 weeks christinapagel.substack.com/p/so-this-is...

We are excited to announce a new, funded PhD studentship opportunity, alongside the School of Informatics @edinburgh-uni.bsky.social! Supervised by @zeerak.bsky.social and starting Sept 2025, the project will examine the ethical implications of natural language processing. Apply ▶️ edin.ac/40PAXEq

Finally got round to migrating my twitter feed to Bluesky... looking forward to reconnecting with folks on here! 👋 Broadly interested in NLP for low-resource languages, specifically for content moderation/LLM bias reduction/creole languages. Get in touch if any of that interests you!

Why is natural language processing technology so much better in English than most other languages? What methods can improve NLP for lower-resource languages? It was great to share my work on this topic at the the @bcs_lovelace consortium and poster competition today! https://t.co/w6VmSG5WL6

Fantastic article on the ethical implications of prompt engineering and shadow prompting by AI companies making large language models. 👏@e_salvaggio from @siegelendowment, published in @techpolicypress https://techpolicy.press/shining-a-light-on-shadow-prompting/

"Whoever controls language models controls politics." Great writing from @hannesbajohr here on the longer-term impacts of gradually replacing human discourse with auto-generated text: https://hannesbajohr.de/en/2023/04/08/whoever-controls-language-models-controls-politics/

Faaaaaascinating. @RishiBommasani, @kevin_klyman, @dzhang105 and @percyliang evaluate how foundation LLM providers measure up to the requirements in the EU's proposed AI Act. Results demonstrate a "striking range" in compliance... @StanfordHAI: https://crfm.stanford.edu/2023/06/15/eu-ai-act.html

Really looking forward to participating in this panel, where we'll take a global tour of diverse approaches to platform regulation and unpack some of the drivers behind them. Join us online *TODAY* at 11.30am CST / 6.30pm BST to hear more. https://t.co/XW8BhcRyJ4 https://t.co/a7ijKBRihz

How do internet shutdowns intersect with violations of political rights and state violence across Africa and the Middle East? If you're at RightsCon in-person, join our dialogue *today* at 12.45 in the Puntarenas room. Details below ⬇️

Really looking forward to contributing to this panel tomorrow and celebrating the excellent research paper produced by @GabeNicholas and @AliyaBhatia at CDT! 10am EDT / 3pm BST, Weds 24 May Sign-up below 👇👇 https://x.com/CenDemTech/status/1661028209901838337

Glad to be in Nairobi for the 2023 Digital Rights and Inclusion Forum, hosted by @ParadigmHQ. Looking forward to our 2 @GlobalPartnersD sessions tomorrow: 1⃣ Government responses to disinfo in SSA 2⃣ Evading accountability for human rights abuses through internet shutdowns https://t.co/ybiApgoGvE

"Large language models and generative AI could be a technology that empowers incumbents much more than it enables disruption." 100% agree with @samgilb on this. At @IntellForum event "Beyond Online Safety: AI, Web3 and the Metaverse." https://t.co/hhopX77hRq

A great pleasure working with you too, @FelAnthonio and @andmeringue! https://x.com/FelAnthonio/status/1636400786874810370

Read my new briefing on a proposed amendment to UK's Online Safety Bill. TLDR: expanding criminal liability for social media managers for content mod duties is a terrible idea. @GOVUK shouldn't try to point-score on online child safety without thinking through the consequences. https://t.co/t3...

The UK's Online Safety Bill is fundamentally flawed in seeking to create an entirely separate online world for children, argue @ellenejudson of @Demos and @kyletaylor of @fairvoteuk. They discuss the risks of such an approach in @politicshome https://t.co/yF6KYth1kl

It was great to chat to @LaVieEnRozena the other week for her article for Welcome to the Jungle on how ChatGPT may shift the landscape of work. https://www.welcometothejungle.com/en/articles/jobs-created-by-chatgpt

Peggy Hicks of @UNHumanRights shares some excellent thoughts on "Defining the way forward" on @UNESCO's guidelines for digital platform regulation #InternetForTrust: 1️⃣ complex problems don't have simple solutions - presenting simple quick fix answers is unhelpful

"Multilingualism on digital platforms is absolutely essential for user empowerment". Dilara Begum speaking on the panel on User empowerment and complaints mechanisms at @unescoNOW's #InternetForTrust with HRH @PrincessRymAli @eSafetyOffice @amalia_toledo @CalGish @DamarJuniarto

From "Amplification and Its Discontents" by @daphnehk @knightcolumbia 👏👏 https://x.com/daphnehk/status/1615020403256164352

Read the latest newsletter by AlgorithmWatch https://r.algorithmwatch.org/nl3/21f8gzjqud4-SLZEdp95Ng

"The intersection between social media and violent extremism remains pervasive and, in fact, a central component of the Internet. Recognizing and confronting the threat posed by this phenomenon is key to preventing another January 6th." https://t.co/LkMEpTVFlC

The UK has amended its Online Safety Bill to remove provisions relating to "legal but harmful" content for adults. How does this and other amendments impact human rights, and how could the bill be further improved? See my analysis for @GlobalPartnersD: https://t.co/0W07D7BtLm

How can we tackle disinformation in a rights respecting way? @SimoneToussi says media literacy initiatives and fact checking are vital but not sufficient to tackle the problem alone - more multistakeholder dialogue is still needed.

At 12:55 GMT, I'll speaking in our #IGF2022 Lightning Talk on govt responses to online disinfo in Sub-Saharan Africa. Come and join to find out what @GlobalPartnersD, @CHR_HumanRights, @cipesaug, @article19wafric and @PROTEGEQV have been working on! https://t.co/fnIdCWE9Az

"The next big change for the internet is to make it accessible regardless of which language and which culture you come from". #ICANN President and CEO Göran Marby at the Opening Ceremony of the #IGF2022 @gcfriyadh, taking place live now

It's great to see @Ofcom proactively investigating technologies used for detection of online harms and pointing out where caution is needed. Their new "Overview of Perceptual Hashing Technology" provides a technical overview and highlights key risks: https://t.co/yzXU53rYo3

"The Online Safety Bill’s drafters should seek to align with the European Union’s Digital Services Act, which recognizes the difference between centralized content moderation carried out by employees, and community-governed content moderation systems." 👏👏 https://t.co/l8eywowx6Q

"Expecting too much from AI or failing to attend to its major shortcomings will lead to continuing disappointment and greater injustice". great & accessible read on why content moderation is so hard and what we stand to gain - and lose - from automating the process👏 https://t.co/TBoigNpFtA

"When it comes to regulating online content we're witnessing some of the other drawbacks of traditional regulation," says @SuzanneNossel of @PENamerica at #OnlinePublicSphere2022. "Regulators don’t understand the platforms well enough to fit them into their paradigms…

Listening to the "Regulating Online Speech: Between the First Amendment and the Digital Services Act" panel at the #OnlinePublicSphere2022. @AgustinaDelCamp begins by asking whether there is consensus around the types of harms we're seeking to address through platform reg.

Kicking off day 2 of the #OnlinePublicSphere2022 conference, @cboteromarino shares 4 key questions for platform regulation: 1⃣What are the responsibilities and rights of platforms? 2⃣ What type of law should regulate content moderation? (@OversightBoard has adopted IHRL)

"One effect of emerging platform regulations is that internal trust conversations within the tech companies are moving out from the purview of the lawyers, and becoming much more integrated into product development teams." @ZoeDarme's opening remarks at #OnlinePublicSphere2022

"The general fallacy of the internet is that it's primarily whatever I don't like about it... but in reality the internet is the nervous system of the 21st century, vital for politics, health, education, employment and everything else important" @doctorow #OnlinePublicSphere2022

Watching the #OnlinePublicSphere2022. @McAndrew kicks off the session on regulatory requirements with 4 key areas of conflict for internet governance: 1⃣ free speech vs. censorship 2⃣ privacy vs. surveillance 3⃣ concentration vs. competition 4⃣ borderless vs. sovereign internet

Did you know that human rights and technology was on the agenda at the 1968 Int'l Conference on Human Rights? Yes, 1968! I certainly didn't know. Fascinating dive into the history of the tech + HR discourse from @StevenLBJensen1 in @OGR_EN https://t.co/kBqgiYPgs8

"To truly, meaningfully understand how machine learning works, there needs to be analysis not just on the model, the data set & the code, but also how it performs when launched & interacting w/ people." @carolinesinders on what transparency really means: https://t.co/VzV6F26fHE

"In the eyes of investors, the balance between maximizing profits and protecting users’ rights is shifting in favor of the latter." fascinating piece by Jan Ryzdak on the value of shareholder activism @ElCalavero @techpolicypress https://techpolicy.press/dissecting-big-techs-shareholder-showdown/

"By empowering content producers that fail the most basic media literacy checks w/ huge audiences, Facebook is exposing communities & users to large risks from bad actors wishing to exploit them for narrow self interest." https://t.co/XKAtjdHTGG