Microsoft is expanding its AI capabilities within Office 365 Copilot by integrating models from Anthropic alongside its existing partnership with OpenAI. Internal testing indicated that Anthropic's Claude Sonnet 4 models demonstrated superior performance in tasks like generating PowerPoint presentations and automating complex Excel functions compared to OpenAI's offerings. This integration aims to enhance the capabilities available to Office 365's over 430 million users. Despite incurring costs for accessing Anthropic's technology through Amazon Web Services, Microsoft plans to maintain the Office 365 Copilot subscription price at $30 per user per month. Meanwhile, the U.S. Senate is examining the White House's AI strategy, with a focus on legislative frameworks to foster long-term AI growth and innovation. In the broader AI landscape, Stability AI has released Stable Audio 2.5 for enterprise use, enabling scalable, high-quality audio generation. Swimlane has also updated its Hero AI platform to enhance security automation with agentic AI capabilities. On the sales front, Vivun has updated its AI Sales Teammate Ava with a new reasoning model and avatar, while Streaml has launched a new AI-powered platform for end-to-end sales automation. Separately, NBA player Tristan Thompson discussed the growth of prediction markets, now a billion-dollar industry, through his role at Tracy AI. A debate is also emerging on whether Artificial General Intelligence should be a global public good, with arguments for and against universal access due to concerns about potential misuse and societal divides. Marketers are being advised to rigorously measure AI investments, focusing on clear benchmarks for productivity, speed, scalability, and output quality, as many current generative AI implementations have shown no clear impact on profits.
Key Takeaways
- Microsoft is integrating Anthropic's AI models into Office 365 Copilot, alongside its existing OpenAI partnership, following internal tests showing Anthropic's superior performance in specific tasks.
- The Office 365 Copilot subscription will remain priced at $30 per user per month, despite Microsoft paying Amazon Web Services for access to Anthropic's AI technology.
- The U.S. Senate Commerce, Science and Transportation Committee is holding hearings to review the White House's AI strategy and discuss legislative frameworks for AI growth.
- Stability AI has launched Stable Audio 2.5, an enterprise-focused audio generation model capable of producing up to three-minute musical compositions quickly and safely for commercial use.
- Swimlane has enhanced its Hero AI platform with agentic AI automation for security operations, allowing natural language commands to trigger automated playbooks and improve incident response.
- Vivun has updated its AI Sales Teammate Ava with a third-generation Sales Reasoning Model and an interactive avatar to enhance B2B sales processes.
- Streaml has introduced an AI-powered platform for comprehensive sales automation, managing tasks from sourcing to scheduling to reduce costs and complexity.
- NBA player Tristan Thompson, involved with Tracy AI, highlighted the significant growth of prediction markets into a billion-dollar industry, expanding into sports.
- A discussion is underway regarding whether Artificial General Intelligence (AGI) should be a global public good, with concerns about equitable access and potential misuse.
- Marketers are urged to establish clear metrics to measure AI investment success, focusing on productivity, speed, and output quality, as many current generative AI uses lack clear profit impact.
Microsoft uses Anthropic AI for Office 365 Copilot
Microsoft is integrating Anthropic's Claude AI models into its Office 365 Copilot, reducing its reliance on OpenAI. This move aims to improve advanced tasks like Excel automation and PowerPoint creation. The company plans to announce this change soon, keeping Copilot priced at $30 per user per month. While OpenAI remains a partner, Anthropic's models were found to perform better in certain areas during internal testing. Microsoft will pay Amazon Web Services to access Anthropic's models.
Microsoft adds Anthropic AI to Office 365 Copilot
Microsoft reportedly plans to use Anthropic's Claude AI models for some Office 365 Copilot features, potentially signaling a shift away from solely relying on OpenAI. Internal tests suggest Anthropic's models create better PowerPoint presentations and handle complex Excel tasks. Microsoft stated its partnership with OpenAI continues, but this move could be influenced by Anthropic's superior performance in specific areas. The Copilot subscription will remain $30 per user per month, despite Microsoft paying Amazon Web Services for access to Anthropic's technology.
Microsoft integrates Anthropic AI into Office 365 Copilot
Microsoft is set to incorporate Anthropic's AI models into its Office 365 Copilot, marking a move beyond its exclusive partnership with OpenAI. Internal testing revealed Anthropic's Claude Sonnet 4 models outperformed OpenAI's in creating PowerPoint presentations and automating Excel functions. This integration blends both companies' technologies within Microsoft's business software, serving over 430 million users. Despite paying Amazon Web Services for Anthropic's models, the Office 365 Copilot subscription will stay at $30 per user per month.
Senate panel reviews White House AI strategy
The Senate Commerce, Science and Transportation Committee is holding a hearing on President Trump's artificial intelligence strategy. The administration has focused on boosting AI technology through various executive orders, including those promoting data centers and tackling 'woke' AI systems. The committee aims to discuss a legislative framework to support long-term AI growth and innovation. Michael Kratsios, director of the White House Office of Science and Technology Policy, is scheduled to testify at the hearing.
Senate holds hearing on Trump's AI strategy
A Senate Commerce, Science, and Transportation Subcommittee convened for a hearing focused on President Trump's artificial intelligence strategy. The hearing addressed the administration's approach to AI development and policy. Further details on the specific discussions and outcomes of the hearing were not provided in the summary.
Marketing AI investment: Measuring success
Marketers must carefully measure their AI investments to ensure they deliver results. Key criteria for evaluating AI solutions include productivity improvement, speed to market, scalability, and quality of output. A recent MIT report found many generative AI implementations had no clear impact on company profits. It's crucial to establish clear benchmarks and track performance beyond just input costs. Focusing on both efficiency and effectiveness is vital for AI to be a true growth investment rather than a marketing expense.
Should Artificial General Intelligence be a free public good?
A debate is ongoing about whether Artificial General Intelligence (AGI), AI that matches human intellect, should be freely accessible to everyone worldwide. Some argue AGI should be a global public good, managed by an entity like the United Nations, ensuring equal access. Others worry that if AGI is costly, it will create a divide between those who can afford it and those who cannot, leading to significant societal disadvantages. Concerns also exist about potential misuse of universally accessible AGI, such as for creating weapons.
Swimlane enhances Hero AI for security automation
Swimlane has updated its Hero AI platform to provide agentic artificial intelligence automation for security operations. The enhanced Hero AI can now leverage Swimlane Turbine's playbooks and integrations in real time, boosting analyst productivity and moving towards autonomous incident response. Security analysts can use natural language commands to trigger automated playbooks, with AI identifying the best actions based on context. The update includes pre-execution summaries and confirmations for user control, ensuring accuracy and trust in automation.
Stability AI releases Stable Audio 2.5 for enterprise
Stability AI has launched Stable Audio 2.5, the first audio generation model designed for enterprise-level sound production. This new version allows for the creation of customizable, high-quality audio at scale, including musical compositions up to three minutes long generated in under two seconds. Stable Audio 2.5 offers enhanced control with audio inpainting and is commercially safe, trained on licensed data. The model can be accessed through Stability AI's website, cloud platforms, or deployed on-premises.
Vivun updates AI Sales Teammate Ava
Vivun has released updates for Ava, its AI Sales Teammate, including a new third-generation Sales Reasoning Model and an interactive avatar interface. Ava uses sales-specific intelligence to understand deals and act proactively, unlike generic AI tools. The new avatar allows real-time collaboration between sales representatives and their AI teammate. Vivun's model is designed for B2B sales environments, aiming to improve contextual reasoning and accuracy to help sellers close more deals.
Tristan Thompson discusses AI ventures and prediction markets
NBA champion Tristan Thompson, who is also the chief content officer at Tracy AI, discussed the growth of prediction markets. He noted that this market has become a billion-dollar industry and is expanding into sports. Thompson shared these insights on Bloomberg's 'The Close,' highlighting the global trend of prediction markets.
Streaml launches AI sales automation platform
Streaml has introduced an AI-powered platform for end-to-end sales automation, managing sourcing, outreach, follow-up, and meeting scheduling. Created by founders frustrated with fragmented sales processes, Streaml consolidates multiple workflows into one system to reduce costs and complexity. The platform offers full-funnel coverage, cross-platform data, intelligent execution by AI agents, and cost efficiency compared to traditional agencies. Streaml aims to help businesses focus on strategy and relationship-building while accelerating sales cycles.
AI investment trend is a decade long, says Adam Parker
Adam Parker from Trivariate believes the current trend in AI investing is a decade-long phenomenon, and we are only two years into it. He also suggested that Oracle's business might have been underestimated by the market. Parker's perspective offers insight into the ongoing AI investment cycle and its future potential.
Sources
- Microsoft Will Use Anthropic Models in Office 365 Copilot
- Microsoft reportedly plans to start using Anthropic models to power some of Office 365's Copilot features
- Microsoft to add Anthropic AI models — as rift with OpenAI grows:...
- Watch live: Senate panel looks at White House AI strategy
- Hearing held on President Trump's artificial intelligence strategy
- Woolley Marketing: Is your marketing AI investment delivering | Analysis
- The Debate On Whether Artificial General Intelligence Should Inevitably Be Declared A Worldwide Public Good With Free Access For All
- Swimlane expands Hero AI to deliver agentic automation for security teams
- Stability AI Introduces Stable Audio 2.5, the First Audio Model Built for Enterprise Sound Production at Scale
- Vivun Introduces Next-Generation Sales Reasoning Model and Avatar Modality for Ava, the AI Sales Teammate
- Tristan Thompson on AI Ventures, Prediction Markets
- Streaml Launches AI-Powered End-to-End Sales Automation
- Two years into a decade-long AI investment trend, says Trivariate's Adam Parker
Comments
Please log in to post a comment.