Get a Free Consult
Strong Tower AI

AI Daily Digest

Your daily briefing on AI developments that matter to your business.

🎧 Listen to this digest

Thursday, September 24, 2026

Top Picks of the Day
Anthropic's Claude Discovers Novel CRISPR-Like Enzyme System in 21 Hours of Autonomous AnalysisGeneral AI

Anthropic announced that its Claude AI model autonomously discovered a previously unknown enzyme system called ART in bacteriophag…

DeepSeek Tops $1 Billion in Annualized Revenue and Targets $7.45 Billion IPO at $75 Billion ValuationInvestment Noteworthy

DeepSeek has reached approximately $1 billion in annualized revenue — a figure that has nearly doubled in recent months after the…

101 AI Language Models Released in 2026 Through September — One Every 2.6 DaysMajor AI Players

A newly published analysis tracking vendor blogs, API changelogs, and press coverage found that 26 labs shipped 101 distinct langu…

Government & Policy

3
See all →
Government & Policy

Supreme Court's Precedents Create Major Legal Hurdles for Federal AI Regulation

A new legal analysis warns that a series of Supreme Court rulings since the 1990s has left the federal government on a very narrow path to regulating AI, particularly open-weight models, at scale. Any regulatory agency created or empowered for this purpose would be presidentially controlled, vulnerable to political shifts, and potentially subject to constitutional challenges around delegation of legislative power. The analysis concludes that Congress would likely need to update any AI legislation frequently — possibly more than once a year — given the pace of model development.

Just Security
Government & Policy

Virginia's New Executive Order Bans NDAs on Data Center Projects, Pushes Transparency and Clean Energy

Virginia Governor Spanberger's Executive Order 22, which took effect September 18, prohibits state agencies from signing nondisclosure agreements that hide material information about data center projects, including public incentives and energy demands. The order replaces former Governor Youngkin's data center policies and establishes a new framework emphasizing transparency, environmental protections, energy costs, and workforce development. Its reach does not yet extend to local governments, with full impact depending on how it is codified into law.

WSLS 10
Government & Policy

Illinois Governor Pritzker Establishes AI Cabinet to Develop Safety and Accountability Regulations

Illinois Governor JB Pritzker signed an executive order creating a new AI Cabinet composed of experts and state agency representatives who will formulate policy proposals through 2027. The unpaid group will analyze AI risks to public safety and critical infrastructure, and will consider tying data center incentives to company safety and transparency standards. The cabinet will also examine how state procurement practices can incorporate AI accountability measures.

My Journal Courier

Major AI Players

2
See all →
Major AI Players

101 AI Language Models Released in 2026 Through September — One Every 2.6 Days

A newly published analysis tracking vendor blogs, API changelogs, and press coverage found that 26 labs shipped 101 distinct language model releases between January 1 and September 23, 2026 — an average of one every 2.6 days. The count excludes minor checkpoints, speech, image, video, and robotics models, and counts model families released on the same day as single entries. September 22 alone saw Claude Opus 5.5 and GPT-6 Sol and Luna launch within approximately an hour of each other.

Medium
Major AI Players

Google Launches Gemini 3.8 Flash TTS: 130-Language Audio Model with Voice Design and SynthID Watermarking

Google released Gemini 3.8 Flash TTS on September 23 as a stable model in the Gemini API and Google AI Studio, offering audio output across 130 languages with voice design, voice replication capabilities, and built-in SynthID watermarking to identify AI-generated audio. The model accepts up to 8,192 text input tokens and is rolling out to Gemini Notebook users, with enterprise API access planned but not yet available. The release marks Google's push to embed production-grade text-to-speech directly into its flagship model ecosystem.

BenchLM

Global Developments

2
See all →
Global Developments

U.S. and China Agree to AI 'Notification System' as Trump-Xi Summit Addresses AI Race

During a high-stakes White House meeting between President Trump and Chinese President Xi Jinping, Treasury Secretary Scott Bessent and China's Vice Premier He Lifeng announced a bilateral AI notification system in which both countries would alert each other if AI goes rogue or causes serious incidents. The agreement comes alongside broader analysis of the U.S.-China AI race, with the U.S. commanding roughly 74% of global AI computing power versus China's 14%, while China leads in research volume and open-weight model releases. A group of Democratic senators separately urged Trump to pursue a more formal agreement to slow frontier AI development.

Al Jazeera
Global Developments

U.S. State Department Orders Diplomats to Replace 'Artificial Intelligence' with 'Super Intelligence' in All Communications

Following President Trump's remarks at the UN General Assembly, the State Department issued a directive ordering diplomats in its international organizations bureau to use the term 'super intelligence' instead of 'artificial intelligence' in all documents and official positions. Trump argued at the UN that 'artificial' makes the technology sound 'fake,' and has pushed the rebrand as part of his broader messaging on U.S. dominance in the sector. The directive represents a concrete administrative implementation of the rhetorical shift announced earlier this week.

AP News

Robotics

3
See all →
Robotics

Chinese Automaker Dongfeng Activates Humanoid Robots in Factories for Sorting and Quality Inspection

Chinese state-owned automaker Dongfeng has begun activating its Xiaodong humanoid robot in production facilities this year, initially handling material sorting and quality inspection tasks. The company aims to bring the robot's efficiency close to a human employee's by end of 2027, with the trial evolving toward more flexible, autonomous operations over time. This joins a broader wave of automakers — including Hyundai with Boston Dynamics — accelerating humanoid robot deployments in manufacturing.

Modern Materials Handling
Robotics

IDTechEx: 80% of Humanoid Robots Are Being Designed for Industrial Use, Market Could Hit 1 Million Units by 2030

A new IDTechEx report finds that four in five humanoid robots currently in development are targeted at industrial manufacturing and logistics applications, with industrial humanoids averaging 78 kg and deployments ramping up from prototypes to early commercial pilots. The firm forecasts the overall humanoid market could approach 1 million units around 2030, though actual deployments have lagged some manufacturers' targets so far. Key remaining challenges include safety certification, training pipelines, and identifying applications that deliver clear, measurable ROI.

ASSEMBLY Magazine
Robotics

O-ID Raises $1.2M Pre-Seed to Build Modular Humanoid Robots for Japanese Manufacturing

O-ID, a Tokyo-based startup founded by two British engineers, has closed a $1.2 million pre-seed round led by TAWANI Ventures to develop fully modular humanoid robots designed for industrial manufacturing and logistics. The company's modular approach is designed to leverage Japan's existing precision manufacturing ecosystem rather than building robot supply chains from scratch. Investors include the early backers of robotics company 1X.

Tech.eu

Transportation

2
See all →
Transportation

Study Finds Automakers Get Better AI ROI from Consumer-Facing Products Than Internal Operations

A new PwC study reported by Automotive News finds that automakers and suppliers are seeing stronger returns on AI investments deployed in customer-facing products than in operational or manufacturing applications. The finding challenges the common assumption that internal efficiency gains are the primary AI value driver in capital-intensive industries. The research comes as automakers face escalating AI investment decisions across everything from in-cabin assistants to autonomous driving systems.

Automotive News
Transportation

Belgian Safety Group Finds Tesla FSD Repeatedly Exceeded Speed Limits and Violated Cyclist Rules in 400km Test

Belgian road-safety advocacy group Johanna.be tested Tesla's Full Self-Driving system over three days in July across approximately 400 km around Brussels, finding it frequently exceeded speed limits in 20 and 30 km/h zones — averaging 44 km/h in 30 km/h areas — and attempted to overtake cyclists where prohibited. The system also incorrectly displayed higher speed limits to drivers, potentially creating dangerous misalignment between what the driver sees and actual road rules. The Flemish transportation ministry described FSD as 'a learning system' that adapts over time.

Fox Business

Healthcare & Biotech

2
See all →
Healthcare & Biotech

FDA Commissioner Nominee Heidi Overton Pledges to 'Go Fast' on Drug Approvals to Compete with China

FDA commissioner nominee Heidi Overton told the Senate that President Trump's directive for her role is to 'go fast,' warning that slow FDA requirements are pushing investment toward China, which she says does things 'faster and cheaper.' She committed to continuing Operation Trailblazer, the administration's initiative to speed up clinical trials, and framed AI-enabled drug development acceleration as central to keeping the U.S. competitive. Her confirmation hearing signals a shift toward faster regulatory timelines for AI-assisted therapeutics.

STAT News
Healthcare & Biotech

FDA Awards Cognita Imaging $1.29M Contract to Test Whether AI Can Evaluate AI Radiology Reports

Cognita Imaging has secured an 18-month, $1.29 million research contract from the FDA to develop and test whether multiple large language models can effectively assess the quality and accuracy of AI-generated radiology reports. The project specifically targets hallucinations — where AI produces plausible but incorrect findings — and omissions of clinically relevant information. The work is part of FDA's regulatory science research into how AI can be validated for clinical use.

MD+DI

Investment Noteworthy

2
See all →
Investment Noteworthy

DeepSeek Tops $1 Billion in Annualized Revenue and Targets $7.45 Billion IPO at $75 Billion Valuation

DeepSeek has reached approximately $1 billion in annualized revenue — a figure that has nearly doubled in recent months after the company raised its model fees by two to four times. The Chinese AI firm is now in the final stages of planning an IPO aimed at raising $7.45 billion, targeting a $75 billion valuation on Shanghai's STAR Market with CITIC Securities as lead underwriter. CEO Liang Wenfeng shared the revenue milestone during recent investor meetings.

Bloomingbit
Investment Noteworthy

Tech Analyst Dan Ives Launches $200M AI-Focused Closed-End Fund, Expected to Price September 29

Prominent tech analyst Dan Ives is launching Ives Ultra AI Opportunities Inc., a closed-end fund targeting approximately $200 million in an IPO expected to price on September 29, with Cohen & Co. managing the deal. Ives, who recently departed Wedbush to launch his own investment bank, Yorkville Ives & Co., will sit on the fund's board through Ives Ultra Capital Management. The fund signals continued institutional appetite for dedicated AI investment vehicles.

Yahoo Finance

General AI

1
See all →
General AI

Anthropic's Claude Discovers Novel CRISPR-Like Enzyme System in 21 Hours of Autonomous Analysis

Anthropic announced that its Claude AI model autonomously discovered a previously unknown enzyme system called ART in bacteriophage DNA, analyzing over 200,000 genes in 21 hours — work that typically takes researchers weeks or months. The system has properties reminiscent of CRISPR and could represent a new gene-editing mechanism with potential applications in gene therapy. The discovery is preliminary and unreviewed, but CEO Dario Amodei cited it at the UN Security Council as evidence that AI is at the 'very beginning' of making medically significant breakthroughs.

TechCrunch
AI
Strong Tower AI
AI Consultant