Hurricane Response, AI Evolution, Community Broadband, and New Tech Initiatives in IT Services

[00:00:02] It's Wednesday, October 9th, 2024, and I'm Dave Sobel. Four things to know today.

[00:00:06] 2Mobile and Starlink launched satellite texting service in Florida as Hurricane Milton approaches.

[00:00:12] 80% of IT workers predict help desk model obsolescence by 2027, driven by AI and automation.

[00:00:21] Anthropics Batch AI API could revolutionize data analysis for mid-sized businesses, offering 50% cost savings.

[00:00:29] And, controlled AI environments emerge as a key use to safe AI usage amid data security concerns.

[00:00:38] This is the Business of Tech.

[00:00:42] With Hurricane Milton hurtling toward Florida, I wanted to share these two stories.

[00:00:47] The FCC has authorized T-Mobile and Starlink to launch their satellite texting service in Florida ahead of Hurricane Milton,

[00:00:55] allowing emergency alerts for all carriers and basic texting for T-Mobile subscribers, even if they lose network connectivity.

[00:01:04] This service is part of a broader response to Hurricane Haleen, with over 10,000 Starlink kits being delivered.

[00:01:11] Users will connect to the service if their phone shows T-Mobile SpaceX and may need to resend message manually if they fail initially.

[00:01:20] I'll also highlight that there is coordination between MSPs, IT service providers and their vendors on Reddit for those who need help, both from Haleen and Milton.

[00:01:31] And while I'm thinking about broadband, the Institute for Local Self-Reliance has released an updated interactive map showing that there are now 400 community-owned broadband networks in the United States, serving over 700 towns and cities.

[00:01:46] Often more efficient and cheaper than private sector options, these networks have been growing rapidly, especially during the COVID-19 pandemic.

[00:01:55] Despite pushback from major ISPs, community broadband initiatives continue to gain popularity and financial support, highlighting a grassroots response to inadequate telecom services.

[00:02:07] And while I'm thinking about sharing, MITRE has launched the AI Incident Sharing Initiative, collaborating with over 15 companies to enhance community knowledge of threats to AI systems.

[00:02:19] This initiative aims to facilitate secure collaboration on incidents involving AI, expanding the MITRE Atlas knowledge base.

[00:02:28] Organizations can submit incidents for data-driven risk intelligence, while the Secure AI project has also updated the Atlas framework to include generative AI threats.

[00:02:39] The initiative seeks to improve collective defense against AI-related incidents across various industries.

[00:02:46] Why do we care?

[00:02:48] Well, tactically, this is on-the-ground information that's quite important for the moment, and I wanted to get that out.

[00:02:54] Now, I want to highlight community-driven efforts.

[00:02:57] Broadband in your community a problem?

[00:02:59] There's a collective solution.

[00:03:01] Security?

[00:03:01] Well, collaboration has long been a staple of that space, and it's hopefully continuing to be true with AI.

[00:03:07] There's opportunity there.

[00:03:12] Thread has declared death to the ticket.

[00:03:15] Did you know tickets date back to the 1800s?

[00:03:18] And customers hate tickets.

[00:03:20] Thread uses an approach for connecting, communicating, and ultimately collaborating.

[00:03:26] Thread allows people to come together around a topic.

[00:03:29] They can discuss and make decisions.

[00:03:31] They can share and invite the right people.

[00:03:34] The future of service is collaborative.

[00:03:37] Supercharge your service experience by seamlessly integrating AI and automation to meet your customers where they are.

[00:03:43] In Teams, Slack, and desktops.

[00:03:46] Not in tickets.

[00:03:47] With instant updates to and from ConnectWise, Autotask, and Halo PSA,

[00:03:52] keep all communication in one place, where it should be.

[00:03:56] With people.

[00:03:57] Decrease time to resolution 30% with chat-based support by visiting getthread.com

[00:04:03] slash MSPRadio to declare death to the ticket.

[00:04:09] A survey by NextThink reveals that 80% of IT workers believe the help desk model will change significantly within three years,

[00:04:18] with many predicting it may become redundant by 2027 due to AI and automation.

[00:04:24] While there's excitement about new technologies, two-thirds of respondents fear for their job prospects.

[00:04:31] The future of help desks may involve evolving into experienced desks, requiring IT staff to upskill in employee experience and generative AI.

[00:04:40] Challenges like employee resistance to self-fixing IT issues and the need for better digital experiences are highlighted as critical for productivity.

[00:04:49] Microsoft has begun allowing partners to sell professional services on its marketplace in the U.S., Canada, and the U.K.,

[00:04:57] enabling solution providers to streamline customer transactions by merging services onto Azure invoices.

[00:05:04] Partners can offer these services as standalone private offers but cannot include them in SaaS offers.

[00:05:10] The initiative aims to enhance the buying experience, although partners have raised questions about transaction handling and its impact on sales goals.

[00:05:19] Microsoft's professional services include assessments, support, and workshops, with custom terms available through private offers.

[00:05:28] Why do we care?

[00:05:29] Two trends here are very related.

[00:05:32] The nature of IT services is going to change substantially over the next few years.

[00:05:36] If 80% of IT workers anticipate a major transformation, businesses should treat this as a clear signal that service models must adapt.

[00:05:45] AI has already shown promise in areas like predictive maintenance, automated ticket resolution, and chatbots for common issues.

[00:05:53] This may mean fewer frontline help desk roles, but the human aspect, such as handling complex multi-system issues or addressing strategic concerns, will remain essential.

[00:06:04] There's also the change in the purchasing process to come.

[00:06:08] Solution providers need to strategize how they will present their services as part of a broader value proposition without the crutch of SaaS bundling.

[00:06:16] For those selling assessments and support, there's an opportunity to differentiate by focusing on customer outcomes and value-added services that integrate with, but don't try directly to, SaaS products.

[00:06:28] The critical takeaway for managed services and solution providers is that the value now lies in offering expertise that combines technology with human-centered services, from managing AI-driven transitions to creating seamless, integrating buying experiences.

[00:06:46] And in the world of AI, Anthropic has launched its Message Batches API, enabling businesses to process up to 10,000 queries asynchronously at half the cost of standard API calls.

[00:06:59] This move aims to make more advanced models much more accessible, particularly for mid-sized businesses.

[00:07:04] The Batch API offers significant cost savings compared to real-time processing, potentially altering how businesses approach data analysis.

[00:07:14] However, it raises concerns about diverting focus from real-time AI advancements, balancing cost with speed in AI implementations.

[00:07:23] And per the information, OpenAI is shifting away from its reliance on Microsoft for data centers due to frustrations over server supply speed.

[00:07:31] Following its significant capital raise, OpenAI is exploring partnerships with Oracle for a powerful AI data center while maintaining a strategic relationship with Microsoft.

[00:07:41] OpenAI is also negotiating terms for its transition to a for-profit model, which may alter Microsoft's stake in the company and its access to future profits.

[00:07:52] Why do we care?

[00:07:54] By processing queries asynchronously, businesses can achieve substantial cost savings.

[00:07:59] This makes AI more accessible to companies priced out of real-time solutions.

[00:08:04] These businesses can leverage AI for larger-scale analysis tasks, such as customer sentiment analysis, bulk data processing, or automated report generation, without the financial burden of constant real-time processing.

[00:08:17] Batch processing might be a way to offer AI-driven services at a lower cost.

[00:08:23] OpenAI is diversifying its supply chain, and that's smart.

[00:08:26] Any changes in this relationship will ripple across the ecosystem.

[00:08:30] Microsoft's Azure offerings, closely linked to OpenAI models, could see future pricing or service changes.

[00:08:37] Providers and partners should monitor this one carefully, as it could impact their AI-based service portfolios and their planning.

[00:08:45] While I usually post big ideas only on Fridays, there are many this week, so I'm choosing one to discuss now.

[00:08:52] From ITPro, CISOs are grappling with the challenge of employees using generative AI tools that risk exposing sensitive company data.

[00:09:00] Despite attempts to implement governance policies, many organizations struggle to enforce them effectively.

[00:09:06] Education and awareness about data risks are emphasized as key strategies to address this.

[00:09:11] Some companies are creating controlled environments for AI to balance regulatory demands and employee needs, allowing safe experimentation with generative AI, while mitigating risks associated with personal tools like ChatGPT.

[00:09:26] Why do we care?

[00:09:28] Employees often bypass official channels to use popular AI tools, even when governance policies exist.

[00:09:34] This is particularly troubling because many of these generative AI platforms, by default, collect data to improve their models.

[00:09:41] Although organizations attempt to create governance frameworks around AI usage, enforcement remains problematic.

[00:09:48] There's a clear need for ongoing education about the risks of using external AI tools.

[00:09:52] It's not enough to tell employees to avoid ChatGPT or similar services.

[00:09:56] There must be a comprehensive understanding of why these tools pose a risk, what happens to the data when entered, and the potential downstream effects of data misuse.

[00:10:07] Creating controlled environments is a vital strategy to balance innovation with security.

[00:10:11] Last week, I highlighted how many organizations will need to do their own internal R&D.

[00:10:16] By providing employees access to internal AI tools or approved platforms within a secure setting, companies allow their workforce to benefit from AI's productivity boosts without risking data breaches, offering that internal R&D space.

[00:10:30] And being that facilitator is a powerful offering.

[00:10:35] Are you and your clients tired of the time-consuming ticket tennis of coordinating meetings and help desk calls?

[00:10:44] Wouldn't it be better to automate this process with a tool that connects directly to ConnectWise Manage or Autotask?

[00:10:53] TimeZest offers scheduling automation that gives you complete control of your schedule and eliminates the hassle of Calendar Paintball.

[00:11:02] As the only service designed specifically for MSPs, it integrates into your workflow and makes scheduling appointments easy on you and your clients.

[00:11:11] Plus, you can try TimeZest for free.

[00:11:16] Visit TimeZest.com slash MSBRadio and use the code MSBRadio to get 10% off your first year of TimeZest.

[00:11:28] Thanks for listening.

[00:11:29] Today is National Pet Obesity Day, so they don't get to participate in International Beer and Pizza Day.

[00:11:35] I'll be doing two live broadcasts as part of S&B Tech Fest on October 17th and 18th, so make sure to sign up and join the event.

[00:11:43] S&BTechFest.com slash go slash Solbel.

[00:11:46] Link in the show notes.

[00:11:47] If you 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.

[00:11:53] And if you enjoy the show, make sure to make sure you're subscribed or following on your favorite platform and tell a colleague.

[00:12:00] We really do appreciate you sharing.

[00:12:03] I'll talk to you again tomorrow.

[00:12:33] Or buy our Why Do We Care merch at businessof.tech.

[00:12:37] Finally, if you're interested in advertising on this show, visit MSBRadio.com slash engage.

[00:12:44] Once again, thanks for listening to me.

[00:12:47] I'll talk to you again on our next episode of The Business of Tech.

[00:12:53] Part of the MSP Radio Network.