AI Fuels Infinite Workdays, Job Cuts at Amazon, and New Trends in IT Service Partnerships

[00:00:02] It's Wednesday, June 18th, 2025, and I'm Dave Sill. Four things to know today. AI is fueling burnout but not solving it, and Microsoft warns that poor time governance is driving an infinite workday. Amazon confirms job cuts tied to AI efficiency gains. Gallup reports AI usage has doubled, but adoption remains uneven and job concerns do linger.

[00:00:24] And TechDial outlines 10 trends redefining the channel, from AI-driven outcomes to shared risk compensation, signaling the end of entitlement-based partnerships. This is the Business of Tech. There's been a lot of discourse about the impact of AI. In a new report, Microsoft warns that the rise of the infinite workday could lead to increased burnout and chaos in the workplace if companies do not change their management of time and priorities.

[00:00:52] The report highlights that 40% of workers check their email by 6am, and meetings after 8pm have risen by 16% year-over-year, indicating a troubling trend where work is encroaching on personal time. The report, part of Microsoft Works Trend Index Special Report, suggests redesigning workflows around artificial intelligence agents and prioritizing high-impact tasks over busy work to alleviate these issues.

[00:01:20] Microsoft also found that nearly 20% of workers check emails on weekends, suggesting that personal time is increasingly being used for work-related tasks. Amazon CEO Andy Jassy just acknowledged that artificial intelligence tools will lead to a reduction in corporate jobs at the company.

[00:01:38] In a company-wide memo, Jassy stated that while the firm will need fewer employees in certain roles due to efficiency gains from AI, it remains uncertain how this will impact the overall workforce in the coming years. Amazon has already experienced a decrease in its overall headcount, with reports indicating that the company employed around 50,000 fewer staff at the end of 2024 compared to its peak in 2021.

[00:02:03] The use of artificial intelligence at the workplace has nearly doubled over the last two years, with 40% of U.S. employees reporting that they have used AI in their roles, up from 21% in 2023. Frequent AI usage, defined as using AI a few times a week or more, has increased from 11% to 19% during the same period, according to Gallup Research.

[00:02:26] The rise in AI adoption is particularly notable among white-collar workers, with 27% indicating they frequently use AI, a significant jump of 12 percentage points since 2024. In contrast, the utilization of AI by product and frontline workers has remained relatively stable. Furthermore, leaders within organizations are more likely to use AI, with 33% reporting frequent use compared to just 16% of individual contributors.

[00:02:54] Despite this growth, employee concerns about job displacement due to AI remain unchanged, with only 15% believing that automation could eliminate their jobs in the next five years. Pax8 has introduced its inaugural research report, the agentic inflection point, which outlines the transformative impact of autonomous software agents on small to mid-sized business operations.

[00:03:17] The report highlights that 54% of mid-sized enterprises have already deployed artificial intelligence, with 83% of high-growth businesses actively experimenting with the technologies. The report identifies four levels of agent functionality, ranging from basic chatbots to coordinated digital workforces showcasing the advancements in automation that are set to redefine industry standards. Why do we care?

[00:03:42] The rise of AI tools was supposed to lighten the load. Instead, they're enabling an always-on culture. Microsoft's findings reveal that AI isn't eliminating busy work, it's just shifting it into personal time. This is not a technology issue. It's a leadership and workflow governance failure. For IT service providers, this is a warning shot. Automating tasks without redesigning how time is managed leads to unsustainable client environments.

[00:04:08] Providers that sell AI services without advising clients on digital well-being, time blocking, and meeting culture are complicit in burning out their customer base. Pax 8's framing of the agentic inflection point and their idea of the rise of the managed intelligence provider suggests a shift from systems management to autonomy orchestration. This is conceptually sound, but currently aspirational.

[00:04:33] Most SMBs are nowhere near agent-level orchestration readiness, and neither are most providers. I continue to argue that the managed intelligence provider idea is limiting and should be rejected nomenclature. IT service providers must read between the lines. AI is not a technical conversation, it's a work or strategy conversation. Simply selling AI tools or productivity software is yesterday's play.

[00:04:59] Tomorrow's winners will be those who help clients redesign time and workflows to protect attention and energy, align headcount strategy with automation opportunities, democratize AI usage across roles, and phase in automation with measurable ROI, not hype cycles. The shift isn't from MSP to AI integrator, it's from tech provider to organizational strategist. Those who don't pivot risk being replaced by those who do.

[00:05:27] Are your customers getting the most from their Microsoft 365 Business Premium subscriptions? Are you delivering maximum value while ensuring best-in-class margins? Nerdio's new modern work features let you streamline the management of Microsoft technologies like Azure Virtual Desktop, Windows 365, Intune, and Defender. Reduce the need for multiple tools, consolidate your vendor stack, and deliver greater value to your customers.

[00:05:54] Help your customers maximize their investments, free up your team for strategic tasks, and drive meaningful business outcomes. With Nerdio Manager for MSP, a single flexible platform with mix and match plans ensures a flexible, perfect fit for you and your customers. Deliver solutions that achieve real business impact. Visit GetNerdio.com to find out more.

[00:06:20] OpenAI has been awarded a $200 million contract by the U.S. Department of Defense to develop prototype artificial intelligence capabilities aimed at addressing critical national security challenges. This contract marks the first collaboration between OpenAI and the Department of Defense, and will last for one year, primarily taking place in the Washington, D.C. area.

[00:06:41] The initiative, named OpenAI for Government, is designed to provide federal, state, and local governments access to OpenAI's advanced models, including specific configurations for national security applications. The Defense Department emphasized that this contract will leverage OpenAI's expertise to enhance various administrative operations, including healthcare for service members and their families, as well as improving data acquisition processes.

[00:07:07] And OpenAI has integrated the Model Context Protocol, or MCP, into ChatGPT, allowing users to connect various tools to the AI, thereby enhancing its capability to access previously siloed data. This move has been supported by major companies, including Google and Microsoft, who have adopted MCP as the emerging standard for data integration with artificial intelligence systems. The integration allows employees to utilize company data in their interactions with ChatGPT and screenwriting workflows.

[00:07:35] However, OpenAI emphasizes caution, as custom MCP servers are not vetted by the company, potentially posing security risks. In other words, users are advised to connect only to trusted servers to prevent issues such as prompt injections or malicious directives that could compromise information security. Why do we care? The DOD awarding OpenAI a $200 million contract is not just a milestone in government tech procurement.

[00:08:03] It's a signal that generative AI is crossing a threshold to new mission-critical infrastructure. The DOD confirms AI is being treated as core infrastructure, not just a business tool. Core business tools trend quickly to commoditization, which should be an additional warning sign. For IT service providers, the window is open to carve out a position as trusted integrators, especially in regulated, public sector, or compliance-heavy environments. But with that opportunity comes responsibility.

[00:08:33] The rise of MCP signals a coming surge of insecure connections and overextended integrations. Vendors and providers who don't lock down their AI workflows will face real-world breaches and policy backlash. Play offense with services that make AI secure, compliant, and connected. That's where the margin lives. Don't just implement AI. Infrastructure it.

[00:08:59] Chinese artificial intelligence startup Minimax has launched its new open-source language model, Minimax M1, which features a groundbreaking context window of 1 million tokens. The model is designed for long context reasoning and has been trained using a highly efficient reinforcement learning technique, significantly reducing operational costs.

[00:09:20] Minimax M1 distinguishes itself with its ability to process extensive amounts of information, outperforming competitors like Odip and AI's GPT-4, which has a context window of 128,000 tokens. The model was developed at a cost of approximately $534,000, compared to other models that have reported training costs exceeding $100 million.

[00:09:41] With 456 billion parameters and two output variants, Minimax M1 offers advanced reasoning and tool use capabilities, making it a competitive option for organizations looking to deploy AI solutions efficiently. Why do we care? Minimax's announcement is less about them and more about the new norm. Massive, open models are cheap, good, and everywhere. That leaves both developers and service providers facing the same question. What's your moat when the model isn't special anymore?

[00:10:11] For IT services firms, the answer is to not wrap the model. It's to own the outcome. Providers who do nothing but deliver model access or lightweight tooling will be swept aside by cheaper, more agile, and more global competitors. The one who survives will build value layers around governance, vertical knowledge, and measurable business results. This isn't the age of models. It's the age of meaningfully applied intelligence. And that's where the channel needs to pivot. Fast.

[00:10:42] And an early big thought. A recent article from TechAisle highlights that the channel landscape is undergoing a significant transformation, driven by emerging trends in technology and changing customer expectations. According to their analysis, 10 core trends are reshaping the channel, and technology vendors, along with their partners, must adapt or risk becoming obsolete. Key trends include the rise of artificial intelligence as a foundational layer for business, which presents built opportunities and challenges for partners.

[00:11:11] Partners are urged to move from reselling products to delivering outcome-based solutions, while embracing shared risk engagement models that align their compensation with client success. The article emphasizes the importance of specialization and real-world project experience as essential factors in establishing authority and gaining customer trust. As traditional marketing development funds decline, vendors are advised to rethink their incentive structures to foster transparent, performance-based partnerships.

[00:11:41] The call to action is clear. Both vendors and partners must innovate and evolve to thrive in this changing environment. Why do we care? The TechAisle piece confirms what's been brewing in the channel. The old playbook is dead. The shift from product-based selling to outcome-based risk-sharing partnerships isn't just theoretical, it's a survival mandate. And for IT service providers and vendors alike, the pressure is structural, not just strategic.

[00:12:08] Quote, move from reselling to delivering outcomes, and quote, is no longer innovative, it's expected. What's new is the shared risk compensation model. Aligning payment to client success, whether uptime, revenue impact, or AI value realized, is a fundamental shift in how value is measured and monetized. With vendors pulling back on traditional marketing development funds, there's a pivot happening. From entitlement-based incentives to performance-based investments.

[00:12:37] MSPs and service firms must get good at attribution, showing vendors exactly how they influence deal velocity or expansion. Without that, they'll be bypassed by more performance-savvy peers. IT service providers must decide, are you a channel partner or a value partner? Thanks for listening. Today is International Sushi Day, International Picnic Day, and National Cherry Tarts Day. Bit of odd picnic.

[00:13:06] Join me for a webinar sponsored by Nerdio, modern endpoint management with Intune. What works and what doesn't. Visit bit.ly slash Nerdio webinar with link in the show notes. A quick programming note, I'll be taking tomorrow and Friday off from regular programming. There's new content coming. In fact, tomorrow my 2025 edition of If I Was Starting an MSP Today episode drops. Expect that one tomorrow. Plus an interview on Friday and the live show in the podcast feed on Saturday.

[00:13:35] Back again on Monday. The Business of Tech is written and produced by me, Dave Sobel, under ethics guidelines posted at businessof.tech. If you've enjoyed the show, make sure you've subscribed or followed on your favorite platform. It's free and helps directly. Give us a review too. If you want to support the show, visit patreon.com slash MSP radio and you'll get access to content early.

[00:14:02] Or buy our why do we care merch at businessof.tech. Have a question you want answered? We take listener questions, send them in, ideally as a voice memo or video to question at MSP radio.com. I answer listener questions live on our Wednesday live show on YouTube and LinkedIn. If you've got a comment or a thought on a story, put it in the comments if you're on YouTube or reach out on LinkedIn if you're listening to the podcast.

[00:14:29] And if you want to advertise on the show, visit MSP radio.com slash engage. Once again, thanks for listening and I will talk to you again on our next episode. Part of the MSP radio network. We'll see you soon. We'll see you soon. Bye.