New ways to buy ChatGPT ads
Monday, May 4, 2026, brings a mixed bag of AI, ML, and data engineering news. From advancements in ChatGPT ads to potential government oversight of AI models, there's a lot to cover.
OpenAI published a case study on its collaboration with PwC to reimagine the office of the CFO, aiming to help enterprises use AI agents to automate finance workflows and improve forecasting. This partnership is built on the idea of leveraging AI to strengthen controls and modernize the CFO function.
On the advertising front, OpenAI expands ChatGPT ads with a beta self-serve Ads Manager, CPC bidding, and enhanced measurement tools, all designed to protect privacy and keep conversations separate from ads. This move marks a significant step in the evolution of AI-driven advertising platforms.
Meanwhile, Databricks posted a blog discussing the foundation of AI scalability, emphasizing the importance of having one team, one platform, and one operating model, particularly in retail where margin pressure is structural. The key takeaway is that companies pulling ahead are those that make faster decisions.
In a series of updates, AWS AI/ML introduced several new features, including the Dataset Q&A feature of Amazon Quick, which powers the next generation of data decisions by allowing business leaders to go beyond traditional BI. This feature is designed to help teams answer unknown questions and make more informed decisions.
AWS AI/ML also introduced the agent quality loop with AgentCore Optimization now in preview, which generates recommendations from production traces, validates them with batch evaluation and A/B testing, and ships with confidence. This is particularly useful because AI agents that perform well at launch often degrade over time as models evolve.
Furthermore, Amazon SageMaker AI now offers an agentic experience that accelerates model customization, allowing developers to describe their use case using natural language and streamlining the entire journey from use case description to deployment. This simplifies the process of building and deploying AI models.
Additionally, AWS AI/ML announced the ability to generate dashboards from natural language prompts in Amazon Quick, significantly reducing the time and effort required to build meaningful dashboards. This feature can generate complete multi-sheet dashboards from simple prompts, making data analysis more accessible.
The company also introduced Amazon S3 Tables as a new data source in Amazon Quick, enabling customers to directly query and visualize Apache Iceberg tables stored in an Amazon S3 table, effectively bridging the gap from data lake to AI-ready analytics.
On the infrastructure side, ClickHouse announced that Gala supercharged its analytics performance with ClickHouse on AWS, improving analytics performance and cutting costs. Gala migrated to the ClickHouse Cloud data platform on AWS to achieve these gains.
In a similar vein, Qonto is rethinking observability with ClickHouse Cloud, replacing sampling and hour-capped queries with two-week query windows, achieving 99.84% compression on high-cardinality data, and leveraging an AI-powered approach to observability at scale.
OpenAI also shared insights into how it delivers low-latency voice AI at scale, detailing its rebuild of the WebRTC stack to power real-time Voice AI with low latency, global scale, and seamless conversational turn-taking. This technical achievement underscores OpenAI's commitment to enhancing user experience through advanced AI technologies.
In funding news, Sierra closed a $950 million round led by Tiger Global and GV, pushing its valuation past $15 billion and giving the company over $1 billion in war chest. The stated goal is to become a leading player in enterprise AI infrastructure.
And one for the security folks: Rossiya Segodnya has deployed a local AI compute cluster built on YADRO GPU servers, keeping sensitive data in-house by moving inference workloads inside its own IT perimeter. This approach allows for the processing of AI workloads while maintaining data security and compliance.
Meanwhile, the US Government is weighing pre-release review for advanced AI models, with the White House exploring formal government oversight of new AI models. The proposed framework would establish a working group to review and approve AI models before they are released to the public.
That's the Daily AI Brief for Monday, May 4, 2026. More tomorrow.
Sources covered
1 story · 14 sources (3 news, 11 company, 0 voices)