Europe provides the AI research, but the capital comes from elsewhere.
Investors in Yann LeCunβs new startup include Jeff Bezos, Eric Schmidt, Nvidia, Samsung and Toyota.
European talent + global capital β a European AI ecosystem.
#EuropeanAI #TechSovereignty #VentureCapital
Still worth a read: Art History and AI: Ten Axioms
The article reflects on how AI can be used responsibly in art historical research and proposes ten guiding principles for integrating machine learning into the discipline.
dahj.org/article/art-...
#DigitalArtHistory #ArtHistory #DigitalHumanities
Claude Opus 4.6 noticed it was being benchmarked, identified the test, found the code on GitHub, and decrypted the answer dataset.
Anthropic disclosed it and adjusted the score.
A reminder: web-enabled LLMs are starting to game benchmarks. οΏΌ
#AI #LLM #AIEvals
Interesting to see publishers using generative AI for scholarly promotion.
Springer automatically generated this short video about my book "Cultural Data Science", which introduces data analysis for art history.
link.springer.com/book/10.1007...
#DigitalHumanities #CulturalDataScience #ArtHistory
The Remote Labor Index (RLI) measures how well AI agents can automate real-world remote work projects β the current best automation rate is just 3.75%.
The debate on AI and the labor market needs empirical indices like the RLI β real-world performance metrics.
#AIandWork #AgenticAI #Automation
Europeβs competitiveness will depend on linking its foundation models with industrial applicability, governance standards, and sectoral integration.
SOOFI: ~100B parameter, reasoning model and agentic applications.
#SovereignAI #EuropeanAI #FoundationModels
Good choice: βslopβ captures not just low-quality output, but the broader erosion of attention, care, and accountability in sociotechnical systems shaped by scale.
Use omni for style transfer
This "labeling / visibility" angle is so important. Tagging @kevinbuist.com + @tanjapraske.bsky.social You might enjoy each otherβs perspectives.
Looping in @rochelle.bsky.social. You two overlap a bit and seem to be circling similar questions.
The University of Virginia Archival AI Protocol, a practical framework for evaluating AI company requests to use cultural heritage collections as training data.
RentAHuman.ai introduces a platform where AI agents can hire humans for real-world tasksβfrom verification and meetings to physical errands.
Are we heading toward the platformization of human presence in service of algorithmic principals?
#AgenticAI #FutureOfWork #HumanInTheLoop
π’ DAHJ is looking for a Community Outreach Coordinator
The International Journal for Digital Art History is seeking an engaged collaborator to help strengthen our scholarly network.
This is a volunteer role for someone excited about #DigitalArtHistory and related fields.
#DigitalHumanities
#Outlier2026 speaker applications are now open! Apply by February 28 for a chance to speak in front of a global audience of data visualizers. Apply now: www.datavisualizationsociety.org/outlier-spea...
Sovereignty is not created by better apps, but by control over defaults, interfaces, and operating models: those who set the defaults shape the market.
#DigitalSovereignty πͺπΊ #PlatformEconomy #TechGovernance
Provenance research needs RAG infrastructures with a human-in-the-loop to ensure scalability, interpretability, and ethical sensitivity.
This paper shows how RAG enables multilingual search in the Getty Provenance Index.
arxiv.org/abs/2508.19093
#DigitalHumanities #ProvenanceResearch #RAG
Cultural Data Science
Cultural Data Science
'Cultural Data Science', by Harald Klinke, addresses growing demand in the visual humanities for foundational knowledge in data analysis, image processing, and artificial intelligence. bit.ly/4ms7FoQ #DataScience #BookSky #DigitalHumanities
What would be your key media studies texts on digital images / digital visual culture and gaming? Not necessarily brand new, rather foundational and still relevant ...
UKRI AI Centre for Doctoral Training in Safe Artificial Intelligence Systems (SAINTS) The SAINTS Centre for Doctoral Training is the UKβs first multidisciplinary PhD programme focused solely on the safety of artificial intelligence (AI).
π£Fully-funded PhD Studentships in Safe AI!
Prompted by further partner support, weβre launching an additional SAINTS recruitment round for PhD students starting in Sept 25. We welcome applicants with backgrounds in Computer Science or Sociology.
Apply πsaints-cdt.ai/2025-studentships/
π #WorldUniversityRankings #LMUMunich #HigherEducation
Wenn sich nun deutsche Plattformen entwickeln (ich glaube es gibt sowas schon von #Siemens? Wer weiΓ mehr?), welche Datenaustausch von spezialisiertem Ingeneurswissen speziell zum Training von #KI / #AI anbieten, dann kΓΆnnte deutsche #IndustrielleKI wieder weltweiter Vorreiter werden.
An understanding of AI as a culturally, ethically, and physically embedded operating system:
Annette Doms argues that Germanyβs strength lies not in raw compute power, but in Industrial AI, system integration, and cultural depth.
#IndustrialAI #DigitalSovereignty #AI2026
Good choice: βslopβ captures not just low-quality output, but the broader erosion of attention, care, and accountability in sociotechnical systems shaped by scale.
Outlier 2026 brings you βThe Final Draft.β #Dataviz is always evolving, & this theme highlights iteration, progress & the power of curiosity, feedback, context, & constraints. Sign up for our newsletter for updates, including the call for speakers & registrations: mailchi.mp/datavisualiz...
Mit Johannes Giesler vom tollen Newsletter @wierechtereden.bsky.social hab ich ΓΌber KI-generierte Bilderwelten als faschistische Wunschmaschinen gesprochen β das GesprΓ€ch kann man jetzt als Podcast anhΓΆren steady.page/de/wie-recht...
2026 is the year AI becomes an infrastructure and organizational problem, not a product one.
The real challenge is no longer models or apps, but compute, energy, governance, and coordination.
Think in stacks, not applications.
ELLIS (European Laboratory for Learning and Intelligent Systems) is building a pan-European, multi-centric AI research network to strengthen Europeβs long-term leadership in AI.
Follow: @ellis.eu
#ELLIS #EuropeanAI #AIResearch #MachineLearning
π Check out the 3 recurring Flagship AI Seminars co-organised by ELLIS Unit Paris offering regular scientific exchanges on key advances in the fields of #AI & #MachineLearning.
π
Ongoing, recurring sessions
π in Paris, sometimes hybrid/virtual
π https://sites.google.com/view/ellis-paris/seminars
