Pressure to Adopt AI Forces MSPs to Absorb Risks Designed by Platform Vendors

[00:00:02] Platform vendors are building AI practice frameworks for MSPs precisely because their own AI economics are under pressure. And MSPs who adopt them are both seeding their methodology and underwriting the vendor's revenue gamble. This is the Business of Tech, I'm Dave Sobel. The channel stack moved at the distribution layer, at the platform layer, and at the partner program level. And across all of it, the direction was the same.

[00:00:30] Start with Pax8 Beyond. The distributor formally launched what it is calling Managed Intelligence, a program and service structure built to give channel partners a defined path for delivering AI services to small and medium-sized businesses. The announcement describes two components, a Managed Intelligence Provider Program, which begins with an AI maturity assessment and moves through role-based training and delivery credentials,

[00:00:56] and managed intelligence services, a delivery layer covering workflow design, Microsoft 365 co-pilot deployment, and AI readiness consulting for S&B clients. Pax8 is anchoring the program in its own research. 62% of S&Bs now consider AI essential to staying competitive, and 74% plan to increase AI spending over the next 12 months.

[00:01:22] The company describes the program as built so that no prior AI expertise is required from the partner to participate. Enrollment is already underway. Pax8 made a second announcement, a partnership with KnockDock structured specifically for MSP's under $5 million in annual revenue.

[00:01:41] That arrangement gives smaller providers access to a 24-hour service desk, a network operations center, and a security operations center without requiring those firms to build any of that infrastructure independently. Juan Fernandez, CEO of Summit Holdings, is quoted in the release saying the arrangement allows sub-scale providers to pursue enterprise-level engagements that were previously out of reach.

[00:02:05] At the distribution layer more broadly, Ingram MicroCloud announced a partnership with PartnerStack to build a new structured channel program around AI services revenue, giving MSPs inside Ingram's network access to partner tooling, deal incentives, and a formalized framework for scaling AI delivery to their client base. ConnectWise fits squarely in the same window. The company launched an AI-native platform built around the concept of predictive IT.

[00:02:32] It combined PSA, RMM security, and agentic automation operations layer, projecting that it will move MSPs from reactive support to autonomous, anticipatory service delivery. If you're listening to this and you haven't hit follow yet on Apple Podcasts, search Business of Tech. It takes five seconds and you'll get the next episode automatically. If you've been watching the backup market, you know pricing has gotten complicated.

[00:03:01] Veeam feels like legacy overhead. Some of the newer platforms have gotten expensive fast. Comet Backup is what I keep seeing MSPs land on when they want modern backup without the modern price tag. Bring your own storage, control your costs, and run it your way. Comet Backup is built for MSPs who want flexibility without the vendor dependency. Check them out at cometbackup.com.

[00:03:28] The SMB online conference is June 23rd through 25th and registration is open now. Three days of practitioner-focused sessions. Pricing, M&A, AI, private equity, service delivery. No vendor speakers, no fluff. The theme is, profitable is enough. If that resonates, you should be there. Small Biz Thoughts community members get in free.

[00:03:53] Everyone else, $399 at smbonlineconference.com. The economics are the starting point. Where's your ed at? laid out what it actually costs to sustain AI at the scale vendors are building toward.

[00:04:10] An estimated 190 gigawatts of new data center capacity at between $9 and $15 trillion to build, requiring roughly $1.75 trillion in annual revenue just to remain viable by 2030. $1.75 trillion a year is nearly twice the entire United States military budget. Not to build something, just to stay solvent on what's already been committed.

[00:04:38] OpenAI's decision to file for an IPO, at a valuation now exceeding $850 billion backed by more than $180 billion in total funding, is that financial pressure made visible. When a company at that scale goes to public markets, it's because the build-out requires a revenue vase that private capital alone can no longer support. The channel is the fastest path to that revenue base at scale.

[00:05:03] And an MSP who adopts a vendor's AI practice program doesn't just become a customer. They become a proof point in a capital market story. The investor case for an $850 billion valuation isn't, we have great technology, it's, we have thousands of partners delivering our methodology to hundreds of thousands of SMBs. The MSP isn't being enabled, the MSP is being enrolled. That revenue has to come from somewhere.

[00:05:31] The Pax8 report on AI and SMB workforce dynamics points at the demand side of that equation. Companies own research found that U.S. labor productivity has risen from 1.43 to 2.16% annually since late 2022. And that generative AI reached 53% population-level adoption within three years, faster than any prior technology wave.

[00:05:57] The headline finding is that AI is breaking the historic link between revenue growth and headcount for small and medium-sized businesses. In plain terms, SMBs are being told to scale with intelligence rather than staff, and most of them have no internal function built to make that work. That gap has a precedent.

[00:06:18] According to Omnia research, roughly 60% of allocated market development funds go unused each quarter, and 43% of partners spend less than half of their allocations. Vendors are committing billions. The research cites Google Cloud, Anthropics, Salesforce, and Dell Technologies. And the channel is consistently leaving it on the table. The money was available. The operational capacity to absorb and deploy it wasn't.

[00:06:45] When capital doesn't convert to capability, and demand doesn't convert to delivery, the gap doesn't stay empty. It becomes someone else's program to fill. The IBM Institute for Business Value surveyed 2,000 senior technology leaders and found that 80% are under active pressure from the CEOs to accelerate AI transformation.

[00:07:09] 11% say they feel prepared for the scale of AI agent deployment expected within the next year. 70% report their organizations are deploying anyway. Strip away the framing and what remains is a behavioral reality. When pressure is high and preparation is absent, organizations do not slow down to build methodology from scratch. They adopt whatever framework is in front of them and move.

[00:07:36] That gap between pressure and readiness is not a waiting room. It's where structured vendor programs enter, get adopted, and get mistaken for a strategy. An MSP that brings a client into a vendor-designed AI practice is answering a genuine need. They are also accepting delivery responsibility for a methodology they did not author.

[00:07:58] When the deployment under delivers, when the workflow doesn't fit the client's operations, when AI spend runs ahead of outcomes, when a compliance question surfaces, the conversation doesn't go to the distributor. It goes to whoever is named in the service agreement. The methodology will belong to the vendor. The accountability will belong to the MSP. That division is not incidental to how these programs are structured. It is the structure.

[00:08:25] The choice is the same one that defines every managed services decision at a structural inflection point. The MSP either becomes the provider that defines what AI does inside a client environment, governs how it operates, and prices that accountability as a service. Or it becomes the delivery vehicle for someone else's framework, absorbing the complexity and the liability without the contract language to be compensated for either.

[00:08:54] This episode is supported by Zero Networks. Cyber resilience is no longer a security team problem. It's a board-level business imperative. When an attacker gets inside a network, the real questions become, how far can they move? Can they get to the crown jewels? How much of the business can they impact and for how long? That's where Zero Networks comes in.

[00:09:16] Zero Networks helps organizations prevent attacks, minimize blast radius, and maintain business continuity, even when attackers get inside. Their micro-segmentation platform automatically builds segmentation policies based on how legitimate users and systems actually communicate, making every access and connection verified and intentional. The result for a threat actor is lateral movement is blocked and threats are contained before they can cause damage.

[00:09:45] Because it's not the breach, it's the damage. Contain the breach before it spreads. The question isn't if attackers gets in, it's whether your business stays running when they do. Zero Networks was built for exactly that. Visit them at ZeroNetworks.com Why do we care? And what does the fair objection look like? The skeptic is right that this isn't new.

[00:10:12] Platform vendors have sold methodology for decades, built peer groups around it, and created whole communities to reinforce it. And the MSPs who committed to those frameworks generally won. The difference with AI is not the methodology. It's where the liability address is. Prior programs told you how to run your practice. You still made every call about what happened inside a client's environment.

[00:10:36] AI methodology tells automated systems what to do inside client environments on an ongoing basis. And when those systems produce a wrong outcome or an unexpected cost, the liability doesn't live at the program level. It lives in the service agreement with whoever signed it. So what to consider? Before enrolling in any vendor AI practice program, separate the offer into two buckets. Delivery tools, the training, credentials, and resources the program provides.

[00:11:06] And methodology, meaning how you assess client readiness, design deployments, and govern AI ongoing. Use the first. Build the second yourself. And keep it in your systems rather than the vendor's portal. Pull your current master services agreement template and look for language covering AI-generated outcomes. Failed workflows, unexpected cost events, compliance exposure from an automated decision.

[00:11:33] If the contract doesn't address those specifically, the default is that you own them. Update that language before you deliver any vendor-designed AI practice to a client, not after the first incident surfaces. The peer group and community model that vendors use to reinforce methodology has genuine value. Shared delivery knowledge, proven practices, faster ramp. Use it for operational learning.

[00:12:00] Just be clear that the community doesn't share your liability when an AI deployment produces an outcome the client didn't expect, and don't let community consensus substitute for your own accountability position. If this trend continues, liability from AI deployments will find its contractual address in the next 12 to 18 months.

[00:12:20] And the MSPs holding agreements built around vendor methodology rather than their own will be accountable for outcomes they didn't design at a price point they didn't negotiate. This is the Business of Tech. Want more from the Business of Tech? Join Business of Tech Plus for ad-free episodes, early interviews, extended cuts, subscriber-only shows, and exclusive member perks and analysis.

[00:12:48] Sign up at businessof.tech.com. And follow this show on your podcast app, and if you're on YouTube, hit subscribe and the bell so you never miss a story. Reviews and comments help spread the word too. Interested in advertising? Head to mspradio.com slash engage. The Business of Tech is written and produced by me, Dave Sobel, under ethics guidelines posted at businessof.tech. Thanks for listening.

[00:13:16] I'll see you on the next episode. Part of the MSP Radio Network.