AI Dispatch: As Global Burns Cash, Local Builds Trust

Here are the top tech stories from our weekly AI news wrap-up (July 17-24).

MITSloan ME Editorial 6 hours ago

Topics

This week, the AI story was as much about governance as it was about the growing pains.

Twenty-nine countries came together to launch the World AI Cooperation Organization. On the security front, OpenAI disclosed that its models — GPT-5.6 Sol and an unreleased, more capable successor — breached Hugging Face’s production systems, reminding the evolving nature of cybersecurity. The spending question loomed large too: Alphabet posted 24% revenue growth on the strength of its AI bets, but AI infrastructure spending pushed free cash flow negative for the first time in over a decade. Closer home, Saudi Arabia’s data authority published a bias reference guide cataloguing over 100 types of AI biases, a new report showed Saudi generative AI adoption more than doubling to 45.2% as internet penetration hit 99.6%, the UAE moved to treat AI literacy as a national economic priority rather than just a training exercise, and Jordan pushed toward full digital government, with 85.5% of public services now online.

1. 29 Countries Agree to Establish World AI Organization

Twenty-nine countries have signed on to launch the World AI Cooperation Organization, a notable move toward coordinated global governance of artificial intelligence. The agreement came just ahead of the 2026 World Artificial Intelligence Conference in Shanghai. It revived a proposal China first introduced at last year’s event, but which failed to secure formal approval.

2. OpenAI Models Breach Hugging Face

OpenAI said two advanced models breached Hugging Face’s production infrastructure during a cybersecurity evaluation.

GPT-5.6 Sol and a more capable pre-release model, operating with reduced safeguards, identified previously unknown vulnerabilities and gained access to external systems. The incident showed that frontier models can conduct complex, multistep cyber operations when normal restrictions are removed.

3. Alphabet Burns Cash for First Time in a Decade Amid AI Spending Surge

Google parent Alphabet on Wednesday reported its second-quarter earnings, announcing that revenue grew by 24% year over year, largely on the back of its AI bets. “Our AI investments are redefining what’s possible across every part of our business,” read the official statement.

While business grew in recent months, spending on AI infrastructure has pushed the tech giant’s cash into negative territory. Its free cash flow, the figure maintained after operations and investments, fell to negative $5.9 billion (£4.3bn) for the first time in at least a decade.

4. Mitsubishi Electric, Sony Plan AI Sensor Venture

Mitsubishi Electric and Sony Semiconductor Solutions will form a joint venture to develop AI-powered vision sensors for factories.

Advanced Vision Solutions is expected to begin operations in October. The venture will combine Sony’s sensor technology with Mitsubishi Electric’s factory-automation systems to automate inspection, production, and other industrial processes.

5. SDAIA Unveils AI Bias Guide, Identifies Over 100 Bias Types

The Saudi Data and Artificial Intelligence Authority has released the first edition of its AI Bias Reference Guide. As AI use grows exponentially across key sectors such as justice, healthcare, and education, the guide identifies over 100 types of biases that can affect the efficiency, accuracy, and fairness of AI systems.

“The complex nature of these systems, which pass through multiple stages involving numerous specialists and experts from diverse backgrounds, increases the likelihood of bias arising during the design and development phases,” the report reads.

6. AI Adoption Doubles as Saudi Internet Penetration Reaches 99.6%

The latest Saudi Internet Report 2025 from the Communications, Space, and Technology Commission (CST) suggests that the Kingdom is entering a phase in which artificial intelligence adoption, digital infrastructure, and online commerce define its digital maturity.

The report shows internet penetration has climbed to 99.6%. AI tool adoption among internet users has more than doubled over the past year, reaching 45.2%, making generative AI a mainstream digital service.

7. UAE Treats AI Literacy as a National Economic Priority: Report

A new report titled Navigating GCC AI Regulation: A Playbook for Building an AI-Literate & Compliant Workforce points out that, rather than treating AI skills as a training initiative, the UAE is positioning workforce capability as a measurable national objective.

It finds that the Gulf’s emerging regulatory model differs from those in Europe and North America by integrating AI literacy directly into governance frameworks, rather than treating it solely as a human capital issue.

8. Jordan Nears Full Digital Government as 85.5% of Public Services Go Online

Jordan’s digital government strategy is moving forward towards redesigning how the state interacts with citizens through AI, automation, and integrated digital infrastructure.

According to Minister of Digital Economy and Entrepreneurship Sami Smeirat, the country has digitized 85.5% of its targeted government services over the past four years and remains on track to bring all eligible public services online by the end of 2026.

Smeirat said the Kingdom has transitioned from fragmented digitization initiatives to a model built around experience and efficiency.

Topics

About the Author

Tags:

AI Alphabet Breach Jordan OpenAI Saudi Arabia Technology UAE

Topics

Share