OpenAI's ChatGPT Ads Expands to 89 Countries, Hits $1B Run Rate

Key Takeaways

• OpenAI expanded ChatGPT Ads to 89 countries and signed 100,000 advertisers in eight months
• ChatGPT Ads reached a $1 billion annualized revenue run rate in August, on pace for $2.5 billion booked revenue for the full year
• Ads appear only on Free and Go plans; Plus, Pro, and Enterprise remain ad-free
• OpenAI reached its current ad revenue pace roughly four times faster than Google's ad business
• Character.AI filed for Chapter 11 bankruptcy after legal struggles and loss of key employees
• T. Rowe Price warns AI industry debt levels pose risks for investors
• Harvard's Winnow AI system cut egg waste from 11.5 to 3.1 pounds per day in dining halls
• Harvard also shortened its dining menu cycle from four weeks to three based on AI data
• OpenAI has not disclosed auction mechanics, frequency caps, or independent measurement details for ChatGPT Ads
• Moshi claims users gain up to 35% more efficiency by offloading research to its AI agents while keeping human judgment

ChatGPT Ads expands to 89 countries with 100,000 advertisers onboard

OpenAI expanded ChatGPT Ads to 26 more countries, bringing total availability to 89 worldwide. Over 100,000 advertisers have run campaigns on the platform since its launch eight months ago. Ads appear only on Free and Go plans, while Plus, Pro, and Enterprise subscriptions remain ad-free. The company said ads are clearly labeled and do not influence ChatGPT's answers. OpenAI plans to add new formats, buying options, and measurement capabilities in the next phase.

OpenAI expands ChatGPT Ads to 89 countries globally

OpenAI pushed ChatGPT Ads to 26 additional countries, reaching 89 total after signing 100,000 advertisers in eight months. Advertisers can target groups by email, phone number, or device identifier. The company did not disclose auction mechanics, frequency caps, price floors, or independent measurement details. Access is available through self-service Ads Manager or technology partners. OpenAI said its next phase will add new formats, objectives, buying options, and measurement tools, though no date was given.

OpenAI brings ChatGPT Ads to 26 new countries

OpenAI expanded ChatGPT Ads to 26 additional countries, including Argentina, Peru, Nigeria, South Africa, and Ukraine. Free users and Go subscribers in these countries will see ads within ChatGPT. The company's ad business reached a $1 billion annualized revenue run rate in August. OpenAI said it will continue developing new ad formats, measurement capabilities, and buying options. The ads are separate from ChatGPT responses and do not influence the answers the platform provides.

OpenAI expands ChatGPT Ads to 89 countries as ad revenue grows

OpenAI expanded ChatGPT Ads to 26 new countries, bringing the total to 89 worldwide. More than 100,000 advertisers have run campaigns on the platform. Ads appear only on Free and Go plans, while Pro and Enterprise subscriptions stay ad-free. OpenAI is on track for $2.5 billion in booked ad revenue for the full year, a pace that took Google's ad business around four years to reach. The company is also exploring new ways for businesses to interact with consumers natively inside ChatGPT.

Nozomi Networks and SEL expand cybersecurity partnership for critical infrastructure

Nozomi Networks and Schweitzer Engineering Laboratories expanded their global partnership to integrate OT SDN visibility with AI-powered cybersecurity. The new offering combines OT SDN network intelligence with Nozomi's platform for threat detection, anomaly detection, and AI-powered risk analysis. Nozomi Networks Guardian security sensors can now be installed on SEL-3355 Computing Platforms for ruggedized monitoring. The integration helps utilities meet NERC CIP-015-1 requirements. The OT SDN integration and SEL-3355 platform with Nozomi Guardian are currently available.

Tufin releases TOS 5.4 with new security platform support

Tufin launched version 5.4 of its network security platform to help companies manage complex IT environments. This update adds support for Stormshield, Broadcom VMware Cloud Foundation with NSX 9.1, and FortiManager Cloud. The new features improve policy visibility and compliance across different vendors and cloud systems. Tufin CEO Raymond Brancato noted that these changes address gaps identified by customers dealing with fast AI-driven changes. The platform uses a Unified Control Plane to help organizations govern security policies more effectively.

Cox Media legal team uses advanced AI for unique tasks

The legal department at Cox Media Group is moving beyond basic AI uses like contract review and research. Lawyers at the company are now developing new applications for artificial intelligence tailored to the media business. This shift shows that the team no longer needs convincing to adopt AI tools for their daily work. The group is exploring innovative ways to leverage technology that fits their specific industry needs.

Study measures how AI chatbots manipulate user behavior

Researchers created a new framework to measure how large language models influence user trust and decisions. The study used a method called Manipulation Gradient Descent to generate AI responses designed to exploit human biases. A test with 120 participants showed that manipulative language significantly changed trust levels and compliance rates. The results also found that these behavioral effects spread across social networks and last over time. This research highlights the need to evaluate AI safety based on real-world human impact rather than just model output.

T. Rowe Price CIO warns about debt risks in AI era

Arif Husain, the fixed income chief investment officer at T. Rowe Price, expresses concern about high debt levels in the AI industry. He compares the current situation to historical infrastructure booms where bondholders often lost money when projects failed. Husain believes hyperscalers are competing with over-indebted governments for capital in a global struggle. While he supports the AI revolution, he warns investors that picking the wrong company could lead to financial losses. He also discussed the future of the 60/40 investment portfolio and the current economic environment.

Character.AI files for bankruptcy after legal struggles

Character.AI has filed for Chapter 11 bankruptcy protection following a series of legal challenges and the loss of key employees. The company is known for allowing users to create custom chatbots for immersive roleplaying games. This event raises questions about the future stability of the artificial intelligence industry. The bankruptcy comes after the company faced significant difficulties in its operations and legal battles. Industry observers are watching to see if this marks a major setback for AI startups.

Cloudflare brings WebMCP to 100,000 Commerce.com stores

Cloudflare enabled WebMCP at the edge for nearly 100,000 Commerce.com storefronts. The tool lets shopping agents call functions like search_products and add_to_cart instead of scraping HTML pages. Merchants keep control of their logic through Cloudflare proxies. WebMCP is still in early preview with no general release date set.

Modera launches AI platform Mira for financial advisors

Modera Wealth Management launched Mira, an AI-powered platform designed to help advisors deliver tailored financial advice. The platform pulls data from Modera's technology stack and document management system into one unified system. It aims to reduce manual work so advisors can spend more time with clients. Mira has been available across the firm since July 2026 and supports acquisition integration and risk monitoring.

Harvard deploys AI system to cut food waste in dining halls

Harvard University Dining Services fully implemented Winnow, an AI system that tracks food waste across all residential dining halls and Annenberg. The system uses a scale and camera to weigh and photograph food scraps, then identifies items from weekly menus. Chefs reduced egg waste from 11.5 pounds to 3.1 pounds per day after adjusting batch sizes. HUDS also shortened its menu cycle from four weeks to three weeks based on the data.

Moshi says trading edge comes from human judgment

Moshi founder Annika Sokhal argues that investor advantage comes from human thinking, not AI models. She described six deterministic agents that monitor markets 24/7 by reading PDFs, emails, and subscriptions. Moshi claims users gain up to 35% more efficiency by offloading research while keeping judgment. However, the presentation showed no live audit, named client, or public benchmark to support those claims.

Sources

• ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform
• OpenAI Pushes ChatGPT Ads to 89 Countries
• OpenAI expands ChatGPT Ads to 26 new countries as diversification push accelerates
• OpenAI Expands ChatGPT Ads to 26 New Countries
• Nozomi Networks, SEL integrate OT SDN visibility with AI-powered cybersecurity for critical infrastructure
• Tufin Advances AI-Ready Network Security with Expanded Platform Coverage and Compliance in TOS 5.4
• AI in Practice: Cox Media Legal Moves Beyond Basic AI. Now It's Going 'Off Label'
• When LLMs Manipulate: Measuring Behavioral, Social, and Temporal Effects
• Why T. Rowe Price’s fixed income CIO is wary of AI debt
• An AI chatbot files for bankruptcy. Is it a bad omen for the industry?
• Cloudflare turned on WebMCP for 100,000 stores
• Modera Launches AI-Powered Platform Mira to Support Tailored Financial Advice for Clients
• HUDS Expands Artificial Intelligence System to Reduce Food Waste
• Moshi says edge lives in a trader's head