Listen to today’s podcast: https://www.youtube.com/channel/UC-nqwUyvLDEvs7bV985k-gQ
AI Daily Podcast — January 10, 2026
Today’s podcast episode was created from the following stories:
$14 billion AI startup Mistral — Europe’s answer to OpenAI — lands French military deal as the region bets on homegrown tech
Source: Original article • By Thibault Spirlet • January 9, 2026
France signed a framework agreement to deploy Mistral AI across its armed forces on sovereign, France-controlled infrastructure. The deal advances Paris’s push for technological and data sovereignty while elevating Mistral as a homegrown alternative to US AI giants. It mirrors rising government adoption of generative AI in defense, with models fine-tuned on sensitive, domain-specific data.
OpenAI、健康特化版｢ChatGPTヘルスケア｣をリリース
Source: Original article • By Ece Yildirim – Gizmodo US（翻訳：岩田リョウコ） • January 9, 2026
OpenAI announced ChatGPT Health, a health-focused experience that can connect to medical records (US-only) and popular wellness apps, with a waitlist open before wider release. The company highlights privacy safeguards such as encrypted, separated health chats that aren’t used to train base models, while EU, Switzerland, and UK availability is limited due to stricter privacy regimes. If successful, it could make AI a more practical companion for understanding test results, prepping for doctor visits, and coordinating diet and exercise.
WEF mapped out 4 AI-driven futures for jobs by 2030, and only one looks good for humanity
Source: Original article • By Thibault Spirlet • January 9, 2026
The World Economic Forum sketches four scenarios for AI and work by 2030, with the “Co-Pilot Economy” offering the least disruption by emphasizing augmentation over replacement. Other paths involve sharper displacement or uneven productivity gains, depending on AI’s pace and how fast institutions reskill workers. The takeaway: outcomes hinge on policy, corporate strategy, and sustained investment in talent readiness.
Boston Dynamics CEO says his humanoid robot will need to be able to learn a new task within 48 hours before it's deployed
Source: Original article • By Lloyd Lee • January 9, 2026
Boston Dynamics aims to have Atlas learn new factory tasks in a day or two as it targets deployment at Hyundai’s Georgia plant by 2028, supported by a partnership with Google DeepMind. The focus starts with logistics tasks, with a long-term goal of 99.9% reliability before tackling more complex assembly. It signals a push toward adaptable, AI-driven “physical” automation on real factory floors.
X makes Grok's AI image tool a premium service after backlash against sexualized deepfakes
Source: Original article • By Robert Scammell and Bryan Metzger • January 9, 2026
Following global backlash and regulatory pressure over nonconsensual, sexualized images generated on X, Grok’s image tool is now restricted to paying subscribers. Lawmakers argue a paywall alone won’t curb abuse, keeping platform liability and safety controls in the spotlight. Expect further scrutiny as regulators probe compliance and victims seek stronger protections.
KI-Update: Grok auf Abwegen, Suizid durch KI?, ChatGPT Health, SleepFM, CES-KI
Source: Original article • By Marko Pauli, The Decoder • January 9, 2026
A European AI roundup highlights mounting pressure on X over Grok deepfakes, OpenAI’s ChatGPT Health rollout with EU exclusions, and a CES spotlight on humanoid robots and powerful new AI chips. It also notes emerging AI commerce standards, Amazon’s controversial “Buy For Me” experiment, and deeper AI features in Gmail. Together, these threads show rapid AI commercialization colliding with safety, data quality, and regulatory realities.
Larian will no longer use GenAI for Divinity concept art, and any GenAI used for other games will be trained on data we own
Source: Original article • By Mark Warren • January 9, 2026
Larian Studios says no generative AI will be used in Divinity’s concept art, and any AI used elsewhere will be trained exclusively on data the studio owns. The team will continue exploring ML for pipelines like animation cleanup and retargeting, again using internal datasets. It’s a notable attempt to balance workflow gains with ethical sourcing and creator consent.
Meta's Reality Labs chief is calling the "most important" meeting of the year, urging employees to show up in person
Source: Original article • By Pranav Dixit • January 9, 2026
CTO Andrew Bosworth called an all-hands for January 14, urging in-person attendance as Reality Labs navigates heavy losses, budget pressure, and Meta’s pivot toward AI. The meeting follows a major AI investment spree and staffing changes alongside cuts within Reality Labs. It signals a pivotal moment for the division’s strategy amid harder questions about product focus and ROI.
