THIS WEEK IN AI AGENTS

The latest launches, strategies, predictions, products & tools

• Microsoft’s 10 new AI agents fortify its grip on enterprise AI

• Snowflake taps Anthropic’s Claude for enterprise and agentic AI offering

• Salesforce launches testing center for AI agents

• H2O.ai boosts AI agent precision with advanced modeling

• Procore unveils AI agents for the construction industry

• xpander.ai’s new step-by-step system makes AI agent more reliable

• OpenText expands capabilities to help enterprises show ROI on AI adoption

• Satya Nadella: Microsoft’s AI agents are only in the ‘middle innings’

• Salesforce, Microsoft and the great AI agent race

• AI agents and simulations are being used to war game crises

• Orchestrator agents: The AI managing the next generation of AI

• Inside Microsoft’s plan to ensure every business has an AI Agent

• Agentic research and the automation of science

• MIT researchers develop novel method to train dependable AI agents

• Google’s new AI Agent Space allows businesses to discover and deploy agents

• Singapore researchers put Anthropic’s ‘Computer Use’ feature to the test

What’s happening in agents right now

How tech giants are racing to define our AI-powered future

The competitive landscape for enterprise AI software shifted dramatically this week, with three tech giants making significant moves that illuminate how businesses will deploy and manage artificial intelligence. Microsoft, Salesforce, and Google have each unveiled ambitious new platforms for AI agents, setting the stage for an intense battle that will shape how companies automate and augment their operations.

The opening salvos

At its annual Ignite conference, Microsoft unleashed an arsenal of 10 new autonomous AI agents designed specifically for enterprise operations. These aren't mere prototypes or proofs of concept - they're fully formed tools backed by an ecosystem of 1,400 third-party connectors and compatibility with over 1,800 language models. The scale of adoption is already notable: 100,000 organizations are actively creating or modifying agents on Microsoft's platform, with deployment rates doubling each quarter.

Salesforce responded with Agentforce Testing Center, a comprehensive platform for evaluating and monitoring AI agents in enterprise settings. The announcement comes after a 30% surge in Salesforce's stock following their launch of Agentforce, prompting Deutsche Bank to raise its target to $365. This isn't just another product launch - it represents the emergence of a new category called "Agent Lifecycle Management," addressing the critical need for AI testing and validation tools in enterprise environments.

Not to be left behind, Google introduced AI Agent Space, an ecosystem program enabling businesses to discover, deploy, and collaborate on AI agents. While currently offering 19 distinct agent models, Google's platform emphasizes partner-driven flexibility and an open ecosystem approach, setting up an interesting contrast with Microsoft's more integrated strategy.

The enterprise AI arms race

This flurry of announcements reflects a deeper truth about where enterprise software is heading. As Jerry Chen articulated in "The New Moats," competitive advantages in technology are increasingly built around systems of intelligence rather than traditional moats like scale or network effects. The tech giants clearly recognize this shift.

Microsoft's strategy, outlined by CEO Satya Nadella, positions AI development in its "middle innings," suggesting significant room for growth and innovation. The company's approach emphasizes practical, manageable, and secure AI applications - a stark contrast to the more experimental approaches seen in consumer AI.

Salesforce CEO Marc Benioff has taken a more combative stance, directly criticizing Microsoft's Copilot offerings while positioning Agentforce as a more comprehensive solution. This tension between the two companies echoes earlier platform wars, but with even higher stakes given AI's potential impact on business processes.

Safety and governance take center stage

The rapid proliferation of AI agents raises critical questions about safety and control. As Sam Harris warned in his analysis of AI risk, the challenge lies in maintaining human oversight while leveraging AI's capabilities. The tech giants appear to be taking these concerns seriously.

Microsoft has integrated robust safety measures through Azure AI Reports and enhanced governance features. Salesforce's Testing Center emphasizes comprehensive monitoring capabilities and sandboxed testing environments. These safety features aren't just add-ons - they're central to the platforms' value propositions.

The new battlefield: integration and implementation

The success of these platforms will ultimately depend on their ability to integrate seamlessly with existing business processes while delivering measurable value. As Clayton Christensen observed in his work on jobs to be done, customers don't buy products - they hire them to perform specific tasks.

This insight is particularly relevant for AI agents. The winners in this space won't necessarily be those with the most advanced technology, but those who best understand and address the actual needs of businesses. Microsoft's extensive enterprise relationships, Salesforce's deep understanding of business processes, and Google's technical prowess in AI each represent different approaches to this challenge.

Looking Ahead: the future of enterprise agents?

Several key questions will shape the evolution of this space:

1. How will pricing models evolve beyond traditional software licensing to reflect the value-based nature of AI agents?

2. What role will open-source alternatives play in challenging the dominant platforms?

3. How will companies balance the promise of automation with the need for human oversight and intervention?

4. Will specialized AI agents emerge for specific industries or functions, or will general-purpose agents dominate?

The enterprise AI agent race is just beginning, but its implications are already clear. As Marc Andreessen argues, AI represents a fundamental shift in how businesses operate. The platforms being built today will likely define how companies function for decades to come.

For business leaders, the message is clear: the future of enterprise software is being shaped now, and the decisions made about AI agent platforms today will have long-lasting implications for organizational capability and competitiveness.

We publish daily research, playbooks, and deep industry data breakdowns. Learn More Here

Agents on the podcast

Our latest conversations around AI agents