The AI Show

Covering the founders, operators & policymakers building AI across Europe & Asia. Live Tues & Thu 11:00 UTC on X + YouTube. Hosts @Diksha_ds @sofiaunchained

[The AI Show @theaisho]

Building AI agents for the messiest workflows in business https://t.co/AyRRQIDLOB

[The AI Show @theaisho]

NEW: Prime Intellect has raised a $130 million Series A at a $1 billion valuation.

[The AI Show @theaisho]

LATEST: OpenAI Launches GPT-Live for natural real-time voice conversations

[The AI Show @theaisho]

https://t.co/pg9D82IOUB

[The AI Show @theaisho]

New: Ukraine's digital ministry says it will favor AI it can run on its own servers and limit models that stay under the provider's control, which by design includes the main OpenAI and Anthropic systems, and the trigger was the US government ordering Anthropic to cut access to some powerful models, which showed Kyiv that a remote model is a tool someone else holds the off switch to.

They are stripping personal data before sending anything to Gemini as an interim fix and building their own model with Kyivstar on Google's open Gemma, due in autumn, for government, business and the military.

[The AI Show @theaisho]

source: https://t.co/T7cOFspyC8

[The AI Show @theaisho]

INSIGHTS: @sysdig documented JADEPUFFER, which they call the first end-to-end ransomware operation run entirely by an AI agent with no human operator behind it, and the detail worth sitting with is that it adapted on the fly, because when its first attempt to create an admin account failed it wrote a corrected payload and retried, going from a failed login to a working fix in 31 seconds.

The real reason this matters is not that AI can now write malware, which we already knew, but that the expensive, patient, human part of an attack, the part where someone sits there for days probing and adjusting can be automated down to seconds.

This kind of ties into a recent update where the ECB gave banks until October to plan for AI-driven cyberattacks.

[The AI Show @theaisho]

Unpacking European Central Bank's request for banks to submit plans addressing AI cybersecurity threats.

[The AI Show @theaisho]

Unpacking European Central Bank's request for banks to submit plans addressing AI cybersecurity threats.

[The AI Show @theaisho]

TAS take on AI & emotional dependency - China's bold move

[The AI Show @theaisho]

https://t.co/Ek0Kiz3rRe

[The AI Show @theaisho]

NEW: Temasek says the 15% is directional, that valuations may get too high to deploy, and that AI has to compete with infrastructure and energy for capital.

[The AI Show @theaisho]

JUST IN: 🇺🇸 US government grants OpenAI approval to fully release its new GPT 5.6 AI model.

[The AI Show @theaisho]

Will frontier models open-source production?

[The AI Show @theaisho]

This Week: OpenAI's Offer, Meta's Pivot, Europe's Alarm https://t.co/EZzI3ef5Ts

[The AI Show @theaisho]

This Thursday, 11AM UTC: @mndl_ai's founding engineer Danitsa Kostova on what it takes to build agents, that handle real supplier chaos. Live on X

tune in!

[Runway @runwayml]

Today we're announcing the opening of Runway's first office in France, establishing a new research hub in Paris anchored by an initial team of ten employees.

France is one of the world's strongest AI ecosystems. Its concentration of world-class research institutions, the depth of engineering talent coming out of them and the French government's sustained investment in AI infrastructure make it one of the most important places to be building frontier AI.

We're hiring in Paris, and across Europe. To learn more and apply for open roles, visit the link below.

[The AI Show @theaisho]

the fable comeback!

[The AI Show @theaisho]

Why are junior devs getting laid off?

[The AI Show @theaisho]

Are token-based AI services stealing your secrets?

[The AI Show @theaisho]

Chatbot export controls and AI geopolitics