April 3, 2026

OpenAI Raises $122 Billion to Build the Infrastructure Layer of the AI Economy



OpenAI announced it has closed a record-breaking funding round of $122 billion at a post-money valuation of $852 billion, positioning itself as core infrastructure for the AI economy. The company emphasizes a reinforcing flywheel driven by consumer adoption, enterprise deployment, developer usage, and access to compute, which together translate technological capability into economic impact. With ChatGPT serving as a massive distribution channel, demand is shifting from basic model access to intelligent systems that reshape business operations, while APIs and tools like Codex expand developer capabilities. The funding is intended to scale research, improve products, expand access, and reduce costs, accelerating AI’s role in productivity, science, and global innovation.

OpenAI highlights exceptional growth metrics, including becoming the fastest platform to reach 10 million, 100 million, and soon 1 billion users, alongside rapid revenue expansion to $2 billion per month. ChatGPT now exceeds 900 million weekly users and 50 million subscribers, dominating consumer AI usage and engagement. Enterprise revenue accounts for over 40% of total revenue and is expected to reach parity with consumer by 2026. Product advancements such as GPT-5.4, Codex expansion, and improvements in memory, search, and multimodal capabilities underpin this growth, while APIs process over 15 billion tokens per minute. Despite strong revenue, the company remains unprofitable and continues to burn cash.

The funding round was supported by major strategic and institutional investors including SoftBank, Amazon, NVIDIA, Microsoft, and firms like Andreessen Horowitz. OpenAI is expanding its infrastructure strategy across multiple cloud providers, chip platforms, and data center partnerships, reinforcing compute as a key strategic advantage. The company is also building a unified AI “superapp” integrating ChatGPT, Codex, and agentic capabilities into a single system to improve usability and drive adoption. At the same time, it faces pressure to justify its valuation ahead of a potential IPO, while managing costs and refining its product portfolio.
Source: WSJ


Waymo’s Robotaxi Surge: Scaling Rides, Markets, and Utilization



Waymo is now delivering 500,000 paid robotaxi rides per week across 10 U.S. cities, reflecting rapid commercial expansion by the Alphabet-owned company. In under two years, weekly rides have increased tenfold – from 50,000 in May 2024 to 500,000 today. Over the same period, Waymo expanded beyond its original markets of Phoenix, San Francisco, and Los Angeles into seven additional Sun Belt cities: Austin, Atlanta, Miami, Dallas, Houston, San Antonio, and Orlando, most of which were added within the past year.

While the company’s fleet has grown, it has remained relatively stable at just over 3,000 robotaxis, according to data shared with regulators in December 2025. This suggests increased utilization per vehicle as ride volumes rise. Waymo may soon update its fleet composition with its 6th-generation self-driving system, set to debut on new vehicles like the Zeekr Ojai minivan and Hyundai Ioniq 5. However, the company faces operational and regulatory challenges, including investigations by federal agencies into robotaxi behaviour near school buses and concerns from San Francisco officials about how Waymo handles stalled vehicles.



Despite its growth, Waymo’s scale remains small compared to Uber’s ride-hailing business, which recorded billions of trips annually. Still, Waymo continues to extend its lead in autonomous ride-hailing. Competitors such as Pony.ai and WeRide operate paid robotaxis outside the U.S., while Tesla has launched a limited paid service in Austin but lacks permits for broader deployment. Other players like Avride, Motional, and Zoox are working toward launching paid services, but they currently trail Waymo’s scale and progress.
Source: TechCrunch


Microsoft Unveils MAI Multimodal Models: Faster, Cheaper, and Built for Real-World AI Deployment

Microsoft announced three new MAI models: MAI-Transcribe-1, MAI-Voice-1, and MAI-Image-2, now available in Microsoft Foundry and MAI Playground. These models emphasize high quality, speed, and competitive pricing. MAI-Transcribe-1 delivers state-of-the-art speech-to-text across 25 languages with a 2.5x speed improvement over Microsoft Azure Fast and strong accuracy benchmarks. MAI-Voice-1 enables natural, expressive voice generation, including custom voice creation from short audio samples, and can generate 60 seconds of audio in one second. MAI-Image-2 enhances image generation with at least 2x faster performance while maintaining quality, targeting professional use cases such as design and marketing.

The models are positioned as cost-efficient and production-ready, with pricing starting at $0.36 per hour for transcription, $22 per 1 million characters for voice, and token-based pricing for image generation. They are already being integrated into Microsoft products like Copilot, Bing, and PowerPoint, and are being adopted by enterprise customers such as WPP for large-scale creative work. Microsoft highlights that these models deliver strong performance in real-world conditions, with MAI-Image-2 particularly noted for natural lighting, accurate textures, and high-quality text rendering in images.

The release reflects Microsoft’s broader strategy to build its own multimodal AI stack while continuing its partnership with OpenAI. Developed by Microsoft’s MAI Superintelligence team led by Mustafa Suleyman, the models are part of a “Humanist AI” approach focused on practical, human-centered applications. Microsoft positions these offerings as cheaper alternatives to competitors like Google and OpenAI, while maintaining enterprise-grade safety, governance, and deployment controls through Foundry.
Source: Microsoft