Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

AI agent projects are just starting and many are barely proof of concepts. What’s clear is your boardroom is about to ask for the agentic AI plan if they haven’t already.

With that backdrop, we dropped in on Boomi World 2025 to get a feel for what enterprises are doing in agentic AI and emerging best practices. Here’s a look at some early lessons to ponder.

It’s early. If Boomi World 2024 was about plotting a course toward agentic AI platforms and enabling enterprises, Boomi World 2025 was about putting the tools in the hands of what Boomi CEO Steve Lucas called “digital alchemists.” Boomi put its Agentstudio in the hands of its customers to build things with a large well of free built-in messages and consumption tiers. It’s obvious that 2026 will be about agents that graduated to production. No matter what the vendor—ServiceNow, Google Cloud, Microsoft, Amazon Web Services, Salesforce and dozens of others—agentic AI projects are just being hatched.

“I don't know if we're a best practices state at this point,” said Boomi Innovation's Michael Bachman in an analyst session. “I think we're going to be at better practices as we iterate on that, but I don't see how we can do it without something like a control tower.”

Boomi World 2025: Boomi acquires Thru, makes case for AI orchestration, automation platform | Boomi World 2025: Agentstudio, AWS pact, 33,000 AI agents deployed | Agentic AI protocols: MCP and A2A today, many more tomorrow

Experiment, but don’t rush in until you’ve sorted use cases and outcomes. Ken Maglio, Principal Architect at World Wide Technology, works in a division of large IT services firm that is spending heavily on centralized AI. As a result, Maglio’s 200-person unit had to wait a bit before investing in its own AI use cases. "We have our own needs but we are not big enough," explained Maglio.

Maglio needed the company's go ahead to invest in AI for its unit and pilots started in the third quarter of 2024. That time frame worked out since Maglio had time to avoid lock-in mistakes by other enterprises. As a result, Maglio looked to build its agentic AI orchestration on Boomi with data and LLMs via Resolve.ai. "The initial questions were how can I use AI to improve experience, drive internal costs down and figure out what agentic looks like a year from now," said Maglio. “We're on our own journey. Basically, they gave us the blessing to go forth and forge our own path and that gave us the time to ask what agentic will look like.”

AI agent projects are more about continuous improvement and options than a destination. Luke Hagstrand, Head of Enterprise AI, Boomi, walked through how Boomi was working through implementing AI agents. The biggest takeaway is that it’s early in agentic deployments and that they are really a continuum from those genAI projects. Boomi began with its ChatB rollout a year ago, focused on sales and marketing for low-risk returns and now is expanding use of AI agents throughout the company.

Lucas said: “I talk about the evolution of the self-driving enterprise. It's not going to happen overnight. This will happen process by process, piece by piece. That's what we want to give organizations. Find the low hanging fruit and deliver trust with that.”

You already have legacy AI technology to worry about. I caught up with two Boomi customers that were focused on data and system integration and already frustrated with early generative AI investments at their companies. These projects worked well enough, but were started more than a year ago. These enterprises, a consumer company and a pharmaceutical vendor, entered deals with OpenAI, which was the only game in town when the projects started. Now they can’t easily go multi-model. Yes, the enterprises have trained OpenAI models on company data and sandboxed them internally. In a nutshell, OpenAI models are the front end interface that taps into the model that has the internal data. “We did it to say we had AI and be on the bandwagon,” said of the customers.

Since you already have legacy AI technology (that’s basically any technology that’s older than 3 months), you’ll have to avoid being locked in. We covered what to look for in an agentic AI vendor last week and horizontal approaches were critical. There are architecture considerations that also matter. For instance, Hagstrand said Boomi took an API-centric approach to generative AI and now agents because it didn’t want to be tethered to any one model or vendor. Boomi uses models from OpenAI, Anthropic and Google Cloud, but can work in more models depending on the use case via Amazon Bedrock. Hagstrand said:

“If we think about this API first, we can try a lot of things and not get locked into large seat based contracts with frontier LLM companies just to try their tools and not even having a baseline for what AI adoption looks like.”

Control the user interface for additional learning and use case knowhow. Boomi’s approach to its internal agents revolved around its ChatB interface. While the models underneath, data and APIs are abstracted away, the data from the interface provides useful information on usage and use cases. “One of the benefits controlling the user interface and having API connectivity is we can track and learn,” said Hagstrand.

There’s a broader point here: Agentic AI means that enterprises can control their own interfaces and more systems are likely to become headless. Boomi CEO Steve Lucas said: “I think SAP will always exist. Workday will always exist; Oracle will always exist. Here's the real question. The real question is, how much of that exists in the future? I believe is their UI will go away entirely.”

Use cases evolve. Those generative AI use cases will come in handy because they create a foundation for agentic efforts. Start with agents that are focused on use cases that can drive measurable returns and then build up to multi-agent systems covering complex workflows. You’re not going to bite off a multi-agent system out of the gate no matter what the vendor tells you. Boomi said 78% of employees now rely on AI agents for daily tasks, productivity has improved 47% and the company has spent 75% less time on customer support requests. Here’s how Boomi’s internal AI agent use cases have evolved.

1. First focus on sales and marketing use cases.
2. Then expand to more deterministic use cases in customer success and support.
3. Create multi-agent systems that handle complex workflows.
4. And then build specialized agents for specific business functions.

Data quality matters. The companies launching genAI and now agentic AI projects thought that they had their data lake houses in order. Once these enterprises scaled, they realize further data work is required to feed the models what they need to deliver accurate answers. Your data hygiene is even more important with multi-agent systems. Some companies are creating data quality agents.

Where data lives also matters. Maglio said where data lives remains a big issue and enterprises need to figure that out first. "The data has to live somewhere," said Maglio, who noted that his division decided to migrate data to Resolve.ai from multiple data repositories. "The data issue is why we slow rolled this," said Maglio. It’s not uncommon for enterprises to have more than a handful of data lakes.

Remember to dust off those old playbooks. A few folks at Boomi World 2025 made the case that agentic AI rhymes with microservices architecture. That take is on target—especially when you consider standards are just being formed and a lot has to be worked out. Microservices took a while to gain traction since tools needed to be built. In the end, AI agents like various microservices have to share data, communicate and ultimately create a modular system that can operate as one (yet be easier to maintain). It’s possible that early agentic AI rhymes with early microservices architecture.

Don’t be scared of consumption models (yet). Hagstrand said Boomi’s internal use of agents is built around a consumption model. The argument here is that the company doesn’t want to pay for seats when not all of them will leverage agents. Agentic AI is not mature enough to make a bet on usage and eating costs on seats.

• Every vendor wants to be your AI agent orchestrator: Here's how you pick
• Agentic AI: Everything that’s still missing to scale
• BT150 zeitgeist: Agentic AI has to be more than an 'API wearing an agent t-shirt'
• Enterprise AI: Here are the trends to know right now
• AI agents bring consumption models to SaaS: Goldilocks or headache?
• Agentic AI: Three themes to watch for 2025

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Alibaba Cloud is seeing more AI workloads, accelerating revenue growth and benefiting from its Qwen open source family of models.

Although Alibaba's overall fourth quarter results didn't impress given tariffs and economic concerns for its retail businesses, Alibaba Cloud saw fourth quarter revenue growth of 18% due to AI workloads. Alibaba Cloud revenue in the fourth quarter was $4.15 billion with operating income of $333 million.

For fiscal 2025, Alibaba Cloud delivered revenue of $16.26 billion, up 11%, with operating income of $1.45 billion.

Speaking on Alibaba's fourth quarter earnings call, CEO Eddie Wu devoted a lot of time to the company's cloud unit. "Revenue from AI related products has maintained triple digit year-over-year growth for the seventh consecutive quarter," said Wu. "We expect AI to remain a key driver of accelerated revenue growth for Alibaba Cloud."

• Alibaba's Qwen picking up momentum
• Google Gemini vs. OpenAI, DeepSeek vs. Qwen: What we're learning from model wars

Wu noted that there are uncertainties in the global AI supply chain, and that customer demand remains strong. AI workloads are a long-term play compared to short-term supply chain fluctuations.

According to Wu, Alibaba will do the following:

• Continue to invest in cloud and AI infrastructure.
• Advance foundational research and innovation in large language models via its Qwen models.
• Leverage open source distribution for Qwen. "By the end of April we had open sourced over 200 models under the Qwen family with more than 300 million downloads worldwide and over 100,000 derivative models, making it the world's largest open source model family," said Wu.

Wu also noted a few trends Alibaba Cloud is seeing. "Among large and mid-sized enterprises, AI applications are expanding from internal systems to more customer-facing use cases. At the same time, adoption of AI product is rapidly extending from large enterprises to a growing number of small and medium-sized businesses," said Wu.

In addition, workloads are broadening. Wu said Alibaba Cloud saw AI workloads in financial services, autonomous driving, internet and online services. Now AI workloads are broadening to more traditional industries, including farming and manufacturing.

"In terms of the trends that we're seeing across these different sectors with more and more companies adopting cloud based AI services, these are companies that had been using a traditional CPU based compute that are now turning to AI and AI compute," said Wu.

Alibaba Cloud's strategy is to leverage Qwen at the edge to drive workloads overall. He said:

"Our open source models have a lot of edge model applications and there are also applications that are suitable -- more suitable to be run on the cloud. There's a lot of different applications. They're not going to have much of an impact in terms of driving cloud business, but because those same customers are using the Qwen models, what that means is often they're also going to require additional usage of cloud based compute resources as well. I think that the edge models to a certain extent are complementary with our cloud based large parameter models. They work well together as a business model."

Constellation Research analyst Holger Mueller said Alibaba is in a good place with its cloud business:

"The four major cloud providers are dukjng it out in each market, but Alibaba has practically a monopoly for workloads in China and for Chinese companies - and doing well accordingly. When is that market saturated? It will be years before we know."

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Salesforce is evolving its Agentforce pricing from a model based on $2 per conversation to a more flexible credit model. Ultimately, customers will be able to transition contracts to incorporate more credits.

The company is rolling out Flex Credits, where pricing adjusts and is based on usage and only charged when an action occurs. Today, Salesforce charges $2 per conversation for all use cases. Going forward, Salesforce will charge by actions, which are defined as specific tasks performed by agents such as updating records and answering questions. There are multiple actions possible within a conversation.

Here's the pricing summary for Flex Credits:

• $500 USD per 100,000 Credits
• One Agentforce action consumes 20 Flex Credits ($0.10 USD)
• All customers with Enterprise Edition or above can get 100,000 Flex Credits for $0 with Salesforce Foundations.
• In the summer, Salesforce will roll out Agentforce user licenses and add-ons for Agentforce for Sales, Service, and Industries, Agentforce 1 Editions for Sales, Service, Field Service, and Industries. Pricing will be announced when generally available.
• Flex Payment Models will be announced in the fall.

Salesforce’s model arrives a few months after the launch of Agentforce, subsequent release and developer and ecosystem follow up and customer feedback on a pricing model that revolved around $2 a conversation.

What Salesforce is trying to do is thread the needle between encouraging Agentforce proof-of-concepts to production, provide one unit of measurement across use cases and futureproof spending on its platform while aligning with value created.

The pricing changes are part of a broader transformation to Salesforce models across its portfolio over the next two to three months. Multiple SaaS vendors are tweaking pricing models to account for AI agents. See: AI agents bring consumption models to SaaS: Goldilocks or headache?

Behind the Scenes: The Force Behind Agentforce | Every vendor wants to be your AI agent orchestrator: Here's how you pick | Agentic AI: Everything that’s still missing to scale

As Salesforce customers evaluate the model, which will enable enterprises to convert seats to credits without early renewal or penalties, there will be multiple moving parts to consider:

• Customers under the Flex Portfolio will need to model multiple plans to optimize. Flex Credits will scale up and down with usage. Editions will be based on seats and Salesforce will launch new ones in June. Agreements will enable enterprises to convert seats to consumption models for users and agents. And there are pay-as-you go plans available.

• Flex Credits will be generally available later this month for Agentforce and cost $500 per 100,000 credits. This model will likely be expanded for other products in the future.
• Flex Credits will include concurrent external and internal agent use cases. The conversation-based model works best for singular use cases.
• Salesforce said Flex Credits will better scale for use cases that deliver value compared to static pricing.
• Flex Credits will allow more transparency and granularity, but also be harder to predict and model initially.
• Salesforce will feature an Agentforce Rate Card that determines how many Flex Credits a customer consumes. These actions will be a mix of included out-of-the-box actions as well as the ability to build custom ones.
• The Einstein 1 Edition will become the Agentforce 1 Edition in the second quarter. For Agentforce, Flex Credits will be included, enterprises can swap seats for agents over time and there are enhanced features and Slack. Einstein 1 was Salesforce's effort to consolidate its products into one suite with Data Cloud and Einstein included.

Flex Credits will be sold under three consumption models including:

• Pre-purchased for the length of a contract and upfront payment with discounts available. This plan is available but doesn't fall under the Flex Credit model.
• Pre-committed, which will be generally available in the third quarter. Pre-committed consumption models mean a customer commits to a contractual amount billed monthly for usage with a shortfall bill if commitment unmet.
• Pay-as-you-Go, which is a Flex Credit plan with no commitment where you're billed on usage. That plan will be available in the third quarter too.

Here are some actions to ponder based on 1 executed action where you'd be charged.

To track and optimize spending, Salesforce is providing a set of tools for transparency and tracking. Salesforce is also developing calculators and simulations to help customers estimate costs under the new model compared to the previous structure.

The company has created a digital wallet for credits where enterprises can configure thresholds, set alerts for stakeholders and proactively monitor usage. Salesforce's Digital Wallet Usage Threshold Alerts are generally available.

Salesforce plans to launch Digital Wallet Usage Tagging, which will provide insights to manage usage based on environment, agent and feature and help customers understand what use case is driving consumption and optimize spending. Usage tagging will be available for Flex Credits in June.

Perhaps the biggest takeaway is that Salesforce is willing to iterate on its pricing as enterprises and vendors work through new AI agent based models. Nevertheless, there will be an adjustment period for the vendor and the customer.

Constellation Research’s take

Salesforce moved to evolve its model after feedback from early customers and deserves props for iterating. However, there is a lot of work ahead for Salesforce.

The biggest mission for Salesforce in this new model is educating customers. It's unclear how this model will impact costs for more complex workflows and use cases.

In addition, transparency will be critical. Tools to project and simulate use cases and associated costs will be critical.

Constellation Research analyst Holger Mueller said:

"Salesforce deserves a lot of credit for being the first vendor who put out a price for the utilization of its agents. But often, early pricing schemes don't stand the test of practicality. With the new flag space pricing, Salesforce moves more into the direction of both outcome and consumption-based pricing.

On the surface, this pricing approach is fairer for enterprises, as outcomes matter to the corporations more than just the invocation of an agent. The pricing model is also fair to Salesforce as well, as pricing needs to reflect the consumed cloud computing resources by agents and therefore needs to be resource consumption based.

Changes to pricing are always a sensitive operation, and we will see how Salesforce will fare here. Only one thing is certain: This will not be the last change of agent pricing in the market."

Liz Miller, an analyst at Constellation Research, said:

“While this move will help address the cost of AI -- a cost that organizations are still struggling to justify and extract maximum value from -- it will also help organizations get started. Right now the reality is that there is a lot of experimentation that falls short of scale.

For its part Salesforce has been focused on getting this ever changing pricing simplified as much as possible. But while Salesforce has looked to perfect a consumption model, others have chosen to build their costs into existing subscriptions noting that AI should just be part of the solution and not an additional feature or additional cost. Which path is right for tech builders and their customers is still up for debate. But the one thing we know for sure: this won't be the last pricing change as AI continues to upend all norms.”

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Cisco reported better-than-expected third quarter earnings as the company saw a surge in AI infrastructure demand and strong results from Splunk.

The company reported third quarter earnings of 62 cents a share on revenue of $14.1 billion, up 11% from a year ago. Non-GAAP earnings were 96 cents a share.

Wall Street was looking for fiscal third quarter non-GAAP earnings of 92 cents a share on revenue of $14.06 billion.

Cisco said product orders were up 20% from a year ago and 9% excluding Splunk. AI infrastructure orders from hyperscale vendors topped $600 million in the quarter.

As for the outlook, Cisco projected fourth quarter revenue of $14.5 billion to $14.7 billion with non-GAAP earnings of 96 cents a share to 98 cents a share. Fiscal 2025 revenue will land between $56.5 billion to $56.7 billon with non-GAAP earnings of $3.77 a share to $3.79 cents a share.

Chuck Robbins, CEO of Cisco, said it was seeing strong demand due to secure networking and AI infrastructure.

Cisco has been stepping up its visibility in emerging technology areas including AI and quantum with a push. Consider:

• Cisco said it will join the AI Infrastructure Partnership (AIP), which is led by BlackRock, Global Infrastructure Partners (GIP), MGX, Microsoft, NVIDIA and xAI. GE Vernova and NextEra Energy also recently joined.
• The networking giant also said it will partner with Saudi Arabia's AI enterprise HUMAIN to scale cost efficient AI infrastructure. Cisco will offer its networking and data center stack, security tools and software.
• Cisco also said it extended a partnership with G42, a UAE-based AI company, to develop AI infrastructure. The two companies signed a memorandum of understanding that revolves around go-to-market, AI infrastructure expertise and global expansion. The two companies will also consider co-developing AI cybersecurity applications.
• Cisco said its quantum networking chip prototype was developed with UC Santa Barbara and generates up to 1 million entangled photo pairs per second at room temperature.

By the numbers:

• Cisco’s networking business posted third quarter revenue of $7.07 billion, up 8% from a year ago.
• Security revenue in the third quarter was up 54% to $2.01 billion.
• Collaboration revenue was up 4% to $1.03 billion.
• Observability revenue was $261 million, up 24%.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Agentic AI protocols have coalesced around Anthropic's Model Context Protocol (MCP) and Google Cloud's Agent2Agent, but Boomi experts expect protocols to emerge to create a crowded field.

Boomi Innovation's Michael Bachman said:

"Now we're somewhere between seven to nine different agent to agent protocols. It seems like we're probably going to have a battle of who's going to win. The nice thing about all of these protocols is we can expose those APIs regardless of whichever type of protocol we're dealing with. It's how we're going to implement those protocols into the stack, platform, runtime, and different modules."

Boomi World 2025: Boomi acquires Thru, makes case for AI orchestration, automation platform | Boomi World 2025: Agentstudio, AWS pact, 33,000 AI agents deployed

Although MCP and A2A have wide support from enterprise vendors, there’s still more work to be done as inter-agent collaboration is sorted out.

Boomi's top protocols include:

• MCP.
• A2A.
• AGNTCY, an open source collective for agent collaboration.

Boomi's plan is to build those protocols into the broader platform. "Each of the modules within the platform itself will have some layer of agentic protocol associated with it in some way, shape or form," said Bachman.

• Microsoft supports Google Cloud's Agent2Agent standard
• OpenAI's support puts MCP in pole position as agentic AI standard
• BT150 CxO zeitgeist: AI agents promising, but in transition period
• Agentic AI: Everything that’s still missing to scale

Bachman said that Boomi will be involved in AGNTCY for agent-to-agent communications. The company will have more to say about that protocol going forward.

Markus Müller of Boomi Innovation said the company is looking to build MCP into APIs and create a gateway that's enterprise grade.

"Just combining what we have with that API to MCP bridge, I think we will really hit the nerve with our customers, making sure they have the API in a shape that is consumable by AI agents and ready to be consumed by either direct access or MCP," said Müller.

For enterprises, MCP developments are like drinking from a firehouse. Müller said that enterprises will need to think through governance, integration and optimization as well as security.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Boomi said it has acquired Thru in a move that will give it the ability to extend its platform into unstructured data. Boomi laid out the case that it should be a platform for AI-driven automation.

Speaking at Boomi World 2025, Boomi CEO Steve Lucas announced the deal. "We've entered into an agreement to acquire a Managed File Transfer platform called Thru," said Lucas. "Thru will give every single one of you the capability to extend Boomi beyond structured data into the world of unstructured."

The extension into unstructured data was part of a broader argument from Lucas that AI agents are a part of a bigger picture that revolves around the intersection of orchestration, automation and integration.

• Boomi World 2025: Agentstudio, AWS pact, 33,000 AI agents deployed
• Agentic AI protocols: MCP and A2A today, many more tomorrow
• Every vendor wants to be your AI agent orchestrator: Here's how you pick

Lucas argued that Boomi customers are integrators that are the "digital alchemists" that power AI transformation. Boomi announced a broader partnership with AWS, availability of Agentstudio and other additions to broaden its platform.

The goal for Boomi: "We are going to commoditize creating AI agents," said Lucas. He walked through Boomi agent building tools for cash to order processes and other workflows. Boomi Agentstudio demos highlighted how agents can find new shipping options and integration with Amazon Q. "We can agentify everything," added Lucas.

He said:

"What is coming is the largest transformation in business, in human not business history. Digital transformation will pale in comparison. What is going to happen? Every software application will be rewritten. The user interface, as we already argued, will change entirely. We won't log into systems, but so much more will be rewritten. Entire software stacks will be built natively around AI. This will all happen. We are not going back."

AI agents have the potential to collapse complexity in businesses and enterprise technology. "What will happen over the next 24 months? It will not take longer, because we will demand it is you'll have an AI agent," said Lucas.

But there's a catch: AI agents are going to add another layer of complexity. Lucas said:

"We have hundreds of apps in our organizations. We have 1000s of databases, 10s of 1000s of APIs. How many agents will we have? If you ask Jensen over at Nvidia, he says 10s of billions with a beat. He didn't even like slow roll that in their line. He was like, well, it could be a lot. He was like, No, 10s of billions. Okay. Well, if there's 10s of billions who's watching them?"

Other takeaways from Lucas' keynote:

Boomi is going with ServiceNow CRM and 'a little petty is fun.' Lucas walked through the ServiceNow partnership. ServiceNow has embedded Boomi into its Workflow Data Fabric. Lucas noted that Boomi is a ServiceNow customer across multiple functions and is now going for ServiceNow CRM. "We're moving to ServiceNow for sales, which means we will be removing gladly, I might add, Salesforce," said Lucas.

AI haves and have nots: "We are in this incredible moment, and I believe that we're seeing this divide, and this divide is something I'm not comfortable with. It's the AI haves and have nots. This is what I worry about. I worry about the small companies, the ones that aren't spending billions of dollars on AI research. Those companies. How will you compete?"

Enterprise software regrets: "I am going to do a little teeny, tiny mea culpa here for a moment, because for 30 years I have participated in this. You got a software thing? I show up. Some of you see me, I'll be like, Hey, man, you want some sap? I got some, right? And Salesforce here, right? We do that. And the pitch, there's always a slide in the deck, and it says, if you just buy this, you will be profoundly happier and super productive, right? It's that is a universal software slide. Everyone has it. It's a thing, and so we buy it. The average company today has 360 SaaS applications alone in the enterprise, 360 so is it 361 that's just gonna fix everything. But we know one universal truth about software is there will always be more of it, and it will be more complex."

All in on AWS: Lucas said Boomi has decided to run solely on AWS because "it gives us the platform that we need to innovate for you."

iPaaS is limiting: "Integration has fundamentally changed. I think about this all the time. What is the word that we call Boomi? Is it integration? Is it iPaaS? I say this with all love and respect to the analyst community, but I bristle a little when I hear iPaaS. It's an if you label me, you negate me kind of a thing," said Lucas, who said integration, automation and AI have blended together.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Boomi said that it has more than 33,000 Boomi AI Agents deployed across enterprises as it rolled out new agents for Boomi Enterprise and Boomi Agentstudio went generally available.

Separately, Boomi and Amazon Web Services announced a strategic partnership that revolves around SAP implementations.

The news, announced at Boomi World in Dallas, comes as the race to become the control tower for enterprise agentic AI heats up. Boomi has many of the characteristics needed to be on a shortlist to manage AI agents.

• Boomi acquires Thru, makes case for AI orchestration, automation platform

• Agentic AI protocols: MCP and A2A today, many more tomorrow

• Every vendor wants to be your AI agent orchestrator: Here's how you pick

Speaking at Boomi World 2025, CEO Steve Lucas said Agentstudio will be built into the platform and available to all customers May 24. Lucas said:

"We want you to leave here excited about creating agents, excited about transforming your processes. You'll be able to build as many agents and test them till your heart’s content. We are also giving you the ability to deploy those agents in a production up to 100,000 messages. If you want to buy more messages with your agent traffic, it's a slider thing. You can just buy more. You can buy less. And then, of course, we have a pro enterprise edition that will allow you to add governance, observability, policies, controls, and audit.”

Ed Macosky, Chief Product and Technology Officer at Boomi, said the company is looking to be positioned at the "intersection of AI, enterprise data, and business process automation."

In a briefing Macosky framed Boomi's opportunity.

"We are under an evolution from being a pure play iPaaS vendor to a broader platform that covers AI, AI agents, automation integration and both data and app integration. We are truly seeing AI agent adoption driven by our customer base. This world of integration, automation and AI is evolving by the minute, but the future is here."

Here's a look at the news:

Boomi Agentstudio, formerly Boomi AI Studio, is generally available. For Boomi, Agentstudio represents its platform strategy to create, manage, govern and orchestrate AI agents. Steve Lucas, CEO of Boomi, said the company is well positioned to be the platform to automate everything with AI agents.

• Agentic AI: Everything that’s still missing to scale
• BT150 zeitgeist: Agentic AI has to be more than an 'API wearing an agent t-shirt'

Agentstudio will include integration with Amazon Q Business. Boomi is also an approved data processor enabled for all AWS customers to aid the building of AI agents on AWS. Boomi also added a new Agent Step feature that allows integration developers to embed registered agents into Agentspace's Process Canvas.

Boomi starts to scale AI agents. The company said customers have deployed more than 33,000 Boomi AI agents to automate tasks, optimize processes and accelerate integration for applications, data and APIs.

• The Future of Agentic AI: Get Ready to Build Your Own Agents With Ease
• Constellation ShortList™ Cross-Platform Agentic AI

The company also outlined new out-of-the-box agents including:

• Integration Advisor Agent, which autonomously reviews integration processes with feedback to improve efficiency and maintenance.
• API Design Agent, which designs and edits APIs and generates OpenAPI specifications.
• API Documentation Agent, which generates business and technical documentation for API definitions.
• Data Connector Agent, which designs and creates data integration connectors for any REST-based data source.

Model Context Protocol (MCP) support. Boomi Enterprise Platform will support native MCP throughout its architecture and API management tools. Boomi Agentstudio will use MCP to access tools via a new gateway for tool aggregation and discovery.

Boomi Data Integration, which was acquired in the Rivery purchase, is now part of the broader Boomi platform. Boomi Data Integration includes automated data pipelines, managed data connectors, change data capture, and more observability tools. Boomi acquires Rivery, adds data integration tools to iPaaS stack

A multi-year AWS collaboration

Boomi and AWS announced a multi-year strategic collaboration agreement that will tie together AI agent management capabilities. The partnership is designed to give customers the ability to build, manage and monitor AI agents across their respective platforms.

In addition, the agreement between Boomi and AWS will focus on accelerating SAP migrations from on-prem to AWS. SAP is also trying to migrate its customer base to the cloud and its S/4HANA platform.

Key items in the Boomi-AWS pact include:

• Amazon Bedrock, which manages models for enterprises, will be integrated with Boomi Agent Control Tower, which centralizes agent management across hybrid and multicloud environments. Boomi Agent Control Tower is part of Boomi Agentstudio.
• Customers will be able to leverage Boomi to manage agents executing in their AWS accounts.
• Boomi's low-code Agent Designer is now integrated with Amazon Q index, which gives AI agents context and model selection.
• Under the agreement with AWS, Boomi launched new native connectors for AWS Lambda, Amazon Bedrock, AmazonDB and Amazon Selling Partner Appstore.
• Boomi for SAP as part of the agreement will provide SAP-certified native integration between SAP and non-SAP systems.
• The companies will support cloud migration extract, load and transform capabilities.
• Enterprises will be able to move SAP data into any AWS data warehouse or data lake.
• Boomi is also an AWS Generative AI Competency Partner.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Databricks said it will acquire Neon, which offers serverless Postgres database services. Databricks' move into the database market is part of its plan to enable agentic AI workloads.

Terms of the deal weren't disclosed, but various reports put the purchase price at about $1 billion.

According to Databricks, more than 80% of databases provisioned on Neon were created by AI agents rather than humans. Databricks is betting that the database market is going to need more scale for AI agents.

• Databricks forges partnership with Anthropic, adds innovative system to enhance open source LLMs
• Databricks, Palantir forge integration pact
• Databricks annual revenue run rate hits $3 billion, compared to Snowflake’s $3.77 billion
• SAP launches Business Data Cloud, partnership with Databricks: Here's what it means

Neon has more than 18,000 customers and competes with AWS Aurora Postgres database service. Microsoft Azure and Google Cloud also have Postgres database services.

Databricks CEO Ali Ghodsi said in a statement that Neon will give developers "a serverless Postgres that can keep up with agentic speed, pay-as-you-go economics and the openness of the Postgres community."

According to Databricks, Neon's serverless Postgres architecture will be integrated with the Databricks Data Intelligence Platform.

This acquisition is just the latest for Databricks, which recently added Fennel. Databricks typically acquires larger startups around its Data + AI Summit in June.

• Databricks adds on to the Lakehouse, acquires MosaicML for $1.3 billion
• Databricks acquires Arcion for $100 million

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

The Stargate Project is currently eyeing as many as three sites for AI data centers and Softbank executives said due diligence is currently underway on multiple proposals.

Speaking on Softbank Group Corp.'s fourth quarter earnings call, a trio of CFOs talked investments in Project Stargate and a portfolio that includes OpenAI, Ampere and Arm.

Project Stargate launched in January with plans to allocate $100 billion to build an AI data center in Texas. Investors included SoftBank, OpenAI, Oracle, and MGX. Technology partners included Arm, Microsoft, NVIDIA, Oracle and OpenAI.

• OpenAI's Stargate Project diversifies its cloud infrastructure
• Softbank buys Ampere for $6.5 billion, says complements Arm, AI infrastructure push
• AI infrastructure becomes one big leveraged bet

Yoshimitsu Goto, Board Director and CFO, said Stargate is an effort in collaboration with chipmakers, energy power producers and power generators. When the first Stargate site launches, that capacity will go to OpenAI. Goto added:

"In the Stargate's program here, right now, there are more than hundreds proposals are being made, and we are having a due diligence for those sites.

But still, not being able to announce first yet. I believe that one will be coming out from state of Texas. Maybe first, second and third will be coming from the state of Texas, but there are several going on parallelly so that we are not yet sure. I cannot really be specific which one will be the first, but I believe we are very close there."

Goto on the earnings call fielded numerous questions about the investment in OpenAI as well as Stargate. Goto said there are plenty of banks willing to finance data centers and tariffs haven't been an issue--largely because Softbank is thinking long-term and tariff volatility and policy changes occur almost daily.

On tariffs and Stargate, Goto said:

"Stargate was announced in January. The tariff discussion had not been started back then. So compared to that time, things are changing every day. At this moment, I don't want to be too specific. We try to explore the best decision, best option at the time. But I don't think the tariff itself is going to stop the project's progress. We would like to wait and see a little bit. But more than that, project itself is something that we would like to make sure to launch, and that negotiation is very important for us."

In many ways, Softbank argued that the acquisition of Ampere, Graphcore and ownership of Arm is all a play to be able to power compute for massive data centers in Stargate. Those three companies in the Softbank family are capable of developing chips for large data centers and there are partners already under contract for other capabilities.

Bottom line: Stargate is playing a big role in Softbank's AI long game. Time will tell how it works out.

Larry Dignan

Editor in Chief of Constellation Insights Constellation Research

About Larry Dignan: Dignan was most recently Celonis Media’s Editor-in-Chief, where he sat at the intersection of media and marketing. He is the former Editor-in-Chief of ZDNet and has covered the technology industry and transformation trends for more than two decades, publishing articles in CNET, Knowledge@Wharton, Wall Street Week, Interactive Week, The New York Times, and Financial Planning. He is also an Adjunct Professor at Temple University and a member of the Advisory Board for The Fox Business School's Institute of Business and Information Technology.&nbsp; <br>Constellation Insights does the following: Cover the buy side and sell side of enterprise tech with news, analysis, profiles, interviews, and event coverage of vendors, as well as Constellation Research's community and… Read more

Asana and Monday.com are seeing customers increase use of AI actions as both companies--along with the likes of Atlassian and Smartsheet--aim to coordinate AI agents and humans.

In recent earnings calls, it's clear that work management platforms are aiming to use agentic AI to be more relevant and coordinate work across platforms.

Monday reported better-than-expected first quarter earnings with revenue growth of 30%. As of the end of the first quarter, Roy Mann, co-CEO of Monday, said users have performed more than 26 million AI actions to date. "We are thrilled to see such rapid growth and usage of AI as our customers utilize the features to automate complex tasks, extract insights and accelerate decision making," said Mann.

Eran Zinman, co-CEO of Monday, said enterprises are the fastest growing market for the company and it launched a bevy of work management features that appeal to enterprises.

Monetization of automated AI actions will come over time, said Mann. "We see a lot of customers get value out of those actions and the numbers are great, but they don't really represent the value," said Mann. "The monetization is still early in early stages and we're experimenting with it. We do see a correlation between usage and pricing and the fact that people do actually pay when they get the real value."

Asana had a similar story and is leaning into AI agents for work coordination. CEO Dustin Moskovitz, who plans to step down, said on the company's fourth quarter earnings call that the goal was to become "the definitive platform for human AI coordination." Asana reports its first quarter results June 3.

• Smartsheet vs. Asana vs. Monday: What you need to know

The company features an AI Studio that now has a Smart Workflow Gallery, a suite of prebuilt AI-powered workflows to improve employee productivity. Smart workflows are focused on marketing, IT and operations. The idea is that Asana can coordinate cross-functional workflows because of its visibility into work and its proprietary Work Graph data model.

"My vision for Asana is to be the defining platform for human-AI coordination. This isn't just about adding AI features. It's about fundamentally transforming how organizations coordinate and execute work at scale, enabling more and higher-level work to become self-driving over time," said Moscovitz.

Asana's fourth quarter revenue was up 10% from a year ago.

Smartsheet is also looking to bring agentic AI to its work management platform via a high-profile partnership with AWS. Specifically, Smartsheet has tight integration with AWS' Amazon Q agentic AI efforts. Smartsheet also launched a new user experience and visualization tools.

The company is likely to have more to say later in 2025 given it has been busy adding leaders since its go-private deal with Blackstone and Vista Equity Partners is now complete. Smartsheet to go private in deal valued at $8.4 billion

All those work management systems are likely to face a threat from Atlassian, which is leveraging its foothold with developers and IT to expand into other team management areas.

Atlassian is developing a System of Work platform that shifts the company from standalone products to a vision of apps and agents grouped into collections. Rovo, Atlassian's AI engine, sits in the middle.

• Atlassian: A look at its system of work strategy, enterprise uptake, AI approach

In the fiscal third quarter, Atlassian delivered revenue of $1.4 billion. The company noted that its Teamwork Graph, which is powered by more than two decades of data on how work is done, can expand its total addressable market.

Atlassian CEO Mike Cannon-Brookes said the company's AI platform has more than 1.5 million monthly active users. "We do think that the future of teamwork is going to be about this sort of iterative human AI human agent collaboration," said Cannon-Brookes. "We've started making the shift from standalone products to a vision of apps and agents with Rovo at the center of everything."