BT 150 interview: Nate Melby, CIO Dairyland Power Cooperative on data, digital twins, smart grids, sustainability

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

Dairyland Power Cooperative, a La Crosse, Wis., is a utility designed to serve rural areas and supply power to customers in four states. It also sits in the middle of multiple trends including energy transition, sustainability and the convergence of information and operational technologies.

I caught up with Nate Melby, Chief Information Officer of Dairyland Power Cooperative, to talk utilities, transformation and IT. Melby is a Business Transformation 150 member.

Transformation and utilities. Melby said utilities are going through a massive transformation that extends from back-office, customer experience to the technology in the field. Operational technology is where the action is. Melby said:

"Not only are we transforming our information technology world, but we’re also transforming those operational technologies. We're in the middle of an energy transition. We're seeing the sources of our generation change and as we add more renewables to the grid, increased complexity creates some new use cases and new scenarios. We have to manage it all and try to do it at scale."

Renewable challenges. "One of the challenges is that the sun doesn't shine all the time, and the wind doesn't blow all the time. And our utilities in the Upper Midwest have extreme weather conditions. Winters are very cold," explained Melby. "How do we continue to sustain the right type of generation for the load that we have across our utility and make sure that we're resilient in the event of those extreme weather conditions?"

With renewables the technology behind the grid has to be smarter and nimbler. The evolution of the smart grid will have to evolve to be more real-time with data about power generation and overall performance. Melby said:

"You had a power plant that would run, and you could control that, but it would run in a steady state. With renewables, you have changes in real time. A cloud goes over a solar array, and the generation drops off for a certain period. We need to be able to control very quickly, and it's not a 15-minute decision in the future, but maybe real time interaction. We see opportunities for machine learning and artificial intelligence to apply that to those problems. There's some potential for us to automate that decision making to provide better control and more resilience."

When will this smart grid arrive? Melby said the smart grid concept is coming in faster than the industry is prepared for just because of demand and need. "We're seeing integration on the power grid before we even have a notion of a virtual power plant and what that could mean. Connecting different types of generation resources in a way we can predict and scale up and down generation with storage and renewables is critical," said Melby.

Digital twins and the grid. Melby said he sees "huge potential for digital twins in our industry." These digital twins could replicate multiple power plants and energy sources so utilities could understand proactive maintenance and the impact of decisions. "Multiply that example by fleet, add the renewables and you can essentially create a virtual model to predict your future performance," he said.

Are the data strategies in place to create virtual power plants? "It's a heavy lift, but one of the advantages of the utility space is that historical data and the capturing of historical data has always been something we've had to do," said Melby, who noted that utilities are regulated to keep 10 years of data. "We have really good foundational data, datasets and data structure. Leveraging that for the future is really about data governance. How do we build the right architecture and the right structure to leverage that data?"

Melby said the historical data at 15-minute and 5-minute intervals are useful, but the future grid will produce "an explosion of data points." "We're seeing exponential growth in data and that means an exponential growth in the data that we have to store. We will need platforms to manage all of that," he said.

Regulation, compliance and security. Melby said regulation, compliance and security has typically meant that utilities kept control of data in closed systems. AS a result, cloud computing has been used on the corporate side but not in operations. Melby said:

"We're now starting to see areas like energy management systems where the cloud is becoming part of it. The cloud architectures have to be built through the lens of managing security and compliance boundaries. Cloud adoption starts with energy management systems, and then distributed energy resource management, and the challenge is how you integrate operations of utilities across regions."

Efficiency and use cases. Melby said utilities are really about efficiency and improving business cases so they can operate at a lower cost and lower rates. "Our mission is to have the lowest rates possible, so we make decisions based on that," he said. Use cases to advance efficiencies revolve around weather prediction, proactive maintenance and managing resources.

Sustainability also overlaps with efficiency. Melby said.

"Sustainability is one of our largest goals. It intersects with technology, efficiency, and the data that we need to effectively manage the grid. Sustainability for us is also about the sustainability of our utility for the people that need us. We provide power in rural areas. We were created to solve the quality-of-life issue where it couldn't be profitable to provide power to rural areas. Our whole mission is to provide power at the lowest cost possible as efficiently and into the future with diverse resources. It's evolving as our industry has been changing, and we're seeing this energy transition happen."

Zoom reports strong Q3, sees AI Companion uptake

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

Zoom Video Communications reported better-than-expected third quarter as the company saw strong usage of its AI capabilities and better retention of small business customers.

The company reported third-quarter earnings of $141.2 million, or 45 cents a share, on revenue of $1.136.7 billion. Non-GAAP earnings in the quarter were $1.29 a share.

Wall Street expected Zoom to report third quarter earnings of $1.07 a share on revenue of $1.12 billion.

CEO Eric Yuan said Zoom's collaboration platform and Zoom AI Companion were showing traction. "We are also pleased with our Online business where we drove higher retention and saw usage of our new AI capabilities, enhancing the value of our platform," he said.

In prepared remarks, Yuan said that Zoom saw 220,000 accounts enabling Zoom AI Companion. Last quarter, Yuan was peppered with questions about generative AI monetization and the company's decision to not charge an add-on price for features. 

• Zoom sets AI Companion roadmap across platform
• Zoom expands reach, generative AI use cases with pricing that aims for market share
• Zoom sees enterprise traction, cites Contact Center, Team Chat wins
• Zoom's customer data terms for training AI may be just the beginning

As for the outlook, Zoom is projecting fourth quarter revenue between $1.125 billion and $1.13 billion with non-GAAP earnings between $1.13 a share and $1.15 a share. For fiscal 2024, Zoom is projecting revenue between $4.506 billion and $4.511 billion with non-GAAP earnings of $4.93 a share and $4.95 a share.

Wall Street was expecting fourth quarter earnings of $1.13 a share and $1.15 a share and fiscal 2024 earnings of $4.66 a share.

By the numbers for the third quarter:

• Enterprise revenue of $660.6 million was up 7.5% from a year ago. Online revenue--the more volatile self-serve accounts--was $476.1 million, down 2.4% from a year ago.
• Zoom Phone reached about 7 million paid seats.
• The number of customers on Zoom One bundles grew 330% from a year ago.
• Zoom Contact Center hit nearly 700 customers at the end of the quarter.
• Zoom ended the quarter with 219,700 enterprise customers, up 5% from a year ago.
• 3,731 customers contributed more than $100,000 in revenue.
• Online average monthly churn was 3% in the quarter, down 10 basis points from a year ago.

Work in a generative AI world will need critical, creative thinking

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

Critical and creative thinking, problem solving, and design are the top skills employers are banking on as generative AI is widely adopted through 2028, according to an Amazon survey.

The survey, which lands roughly a week ahead of AWS' re:Invent conference, was based on 1,340 employers and 3,297 employees in the US.

Amazon outlined some key takeaways about usage--92% of respondents expect to use AI applications by 2028 with IT being the biggest beneficiary--but many of those items were known. Yes, we know generative AI will have a huge impact on multiple departments and routine tasks will be automated.

What's left out of many of these generative AI discussions is what skills will be necessary in the new work landscape. When you consider two years ago folks were preaching coding to kids. Today, generative AI does a lot of the heavy lifting. With that backdrop, Amazon's more detailed data in its PDF caught my eye.

In the report, Amazon wrote:

"Critical thinking is essential to evaluate the accuracy and relevance of AI outputs, while problem-solving helps optimize the capabilities of AI systems by defining and structuring analyses appropriately on available data. Ethics and risk management is also ranked as the fourth most important skill needed to use AI effectively. That’s because while AI can mimic many human skills and competencies, it still falls short in other areas, like emotional intelligence, contextual understanding, common sense, adaptability, ethics, and intuition."

Here's the data from the survey. More than half of respondents say thinking well will be key to using AI well. Technical skills were cited by 47%.

Also see: How Generative AI Has Supercharged the Future of Work | Generative AI articles | Why you need a Chief AI Officer | Software development becomes generative AI's flagship use case | Enterprises seeing savings, productivity gains from generative AI 

Not surprisingly, a lack of skills, knowledge and career paths were cited as barriers to AI adoption.

Other key items from the report include:

• The interest in AI skills crosses multiple generations with GenZ and Millennial employees interested in developing AI skills checking in at 81% to 84% with Gen X at 78%. Indeed, 65% of Boomers were interested in AI skills too.
• 47% of respondents said IT was the department expected to benefit the most from AI skills, followed by sales and marketing, finance and business operations.
• 62% of employers expect generative AI to boost innovation and creativity, but only 52% of employees do. The two sides were roughly aligned on automating tasks. 

Microsoft, OpenAI, Altman and the enterprise: What did we learn?

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

In a boardroom drama designed for the TikTok generation, Sam Altman was ousted as CEO of OpenAI, negotiated for a return and then landed at Microsoft along with Greg Brockman to lead an "advanced AI research team." The fact Microsoft CEO Satya Nadella moved so quickly illustrates the high stakes.

Nadella said on X that it will remain a partner of OpenAI, led CEO Emmett Shear, Co-founder of Twitch, but added Altman and Brockman. Perhaps the biggest takeaway--at least for Microsoft's market cap--is that Nadella said:

"We have confidence in our product roadmap, our ability to continue to innovate with everything we announced at Microsoft Ignite, and in continuing to support our customers and partners."

For enterprises, that's about all you need to know about this OpenAI soap opera. A brief recap:

• OpenAI ousts Sam Altman as CEO over lying to board
• Altman is back, not back, lands at Microsoft in what could be a Steve Jobs-ish moment
• Nadella hearts OpenAI, but really hearts Altman and Brockman on X
• A roundup of news coverage via Techmeme

Given that OpenAI was the hub for a good bit of Microsoft's Copilot strategy, the software giant needed to manage through the OpenAI-Altman drama quickly and possibly use its own Azure models-as-a-service. On Friday, OpenAI said it fired Altman. Brockman resigned after the news. Key executives at OpenAI also started to bail.

Altman said on X that the "mission continues." The back-and-forth with Altman, Nadella and Brockman was only interrupted by Tesla and X CEO Elon Musk noting that Altman will have to use Teams now (Altman was fired on Google Meet).

Constellation Research CEO Ray Wang raised the questions about who will acquire OpenAI, where talent goes and the economics of generative AI. Here are a few other observations to ponder.

• Microsoft and OpenAI were a perfect match between a massive company with resources and a startup that could innovate quickly--until it wasn't. This blueprint will be analyzed going forward as a strategy case study. Did too much of Microsoft's generative AI strategy hitched to just a few people that didn't work for the company?
• Diversification matters. As I noted before, there can be too much choice in large language models. The trick is navigating how much generative AI choice gives an enterprise diversification. Bring your own model will look much better after this OpenAI fiasco.
• Yes, you'll need a Chief AI Officer to sort vendor AI messes.
• Own your intellectual property. The Microsoft-OpenAI weekend illustrates why your IP needs to be owned by you even if you're building on top of an LLM. The model that may emerge is one that revolves around first party data and open-source models.
• Don't bet too much of your strategy on a startup. Yes, OpenAI was an odd duck with a board composed of non-profit types and venture capital types. Startups can give enterprises more innovation and attention but have a plan in case of an emergency.

Retailers hope business, digital transformation efforts pay off for holiday shopping 2023

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

With holiday shopping season underway, retailers see uncertain consumer spending, better inventory positions, lower supply chain costs and customer experience investments they hope deliver.

Adobe is forecasting US online holiday sales of $221.8 billion (Nov. 1 through Dec. 31) in 2023. Salesforce noted that 2023 is all about keeping loyal customers happy and focusing on the experience metrics that matter.

For our purposes, retail in 2023 matters because the industry is arguably the best lab for digital and business transformation, analytics, customer experience and a bevy of technologies.

Here's a tour of how retailers are setting up for the holidays.

Walmart

Walmart's results are the barometer for consumer health and digital transformation.

"In the U.S., we may be managing through a period of deflation in the months to come. And while that would put more unit pressure on us, we welcome it because it's better for our customers," said Walmart CEO Doug McMillon.

Those comments rattled the stock market. CFO John Rainey added that consumers are making trade-offs and weekly performance metrics were softening at the end of October.

Like other retailers, however, Walmart is focusing on what it can control across its omnichannel experience. McMillion said:

"We're making shopping easier and more convenient. Our net promoter scores for pickup and delivery in Walmart U.S. are improving and we've started using generative AI to improve our search and chat experience. We've released an improved beta version of search to some of our customers who are using our app on iOS. In the coming weeks and months, we will enhance this experience and roll it out to more customers."

Rainey said omni-channel services such as pickup and store fulfilled delivery are driving growth. Walmart 24% e-commerce revenue growth in the third quarter and share gains among higher income households.

On the digital front, McMillon said Walmart is looking to blend its core retail business with new services including membership, third-party marketplaces, and advertising. The combination of these businesses with automation in the supply chain should provide a good mix of operating margins, growth and predictability.

• Walmart, Target highlight intersection of supply chain, customer experience
• Walmart Chief Sustainability Officer: ESG is integrated into operations, business
• Amazon vs. Walmart: 8 innovation takeaways

Retail experiences all start at the back end. To that end, Rainey said:

"During the quarter, we opened our third next-generation e-commerce fulfillment center. These 1.5 million square feet facilities are expected to more than double the storage capacity, enable 2X the number of customer orders fulfilled daily, and will expand next and two-day shipping to nearly 90% of the U.S. including marketplace items shipped by Walmart Fulfillment Services. They also unlock new opportunities for our associates to transition into higher skilled tech focused positions."

Macy's

Macy's CEO-Elect Tony Spring said its customers at Macy's, Bloomingdales and BlueMercury will continue "to be under pressure and discerning and how they spend in discretionary categories we offer." He said Macy's is ready to fulfil orders online, in store and through its gift guides.

Spring also argued that the company's department stores can pivot with content and merchandise that goes where customers are headed.

On the digital front, Spring said that Macy's transformation is on track. The company is looking to run three distinct retailers with their own identities and a common data and technology approach. Spring said:

"We can learn from each other without becoming one another as we remove silos to optimize our collective customer insights.

We are also balancing art and science. I like to say that this is STEAM, not STEM. We are embracing data science tools, including AI and machine learning, to drive more accurate and agile decision-making based on changes in demand. This, married with the art of human judgment, helps us become more proactive and customer influenced."

Spring said Macy's will leverage data to create better experiences and scale its growth vectors. The growth vectors include private brands, small format stores, a digital marketplace, a focus on luxury and personalized offers and communications.

Macy's CFO and COO Adrian Mitchell said the company's supply chain is working well and the retailer has a good inventory position that's down 6% from a year ago and down 17% from 2019 levels. The company is seeing lower freight expenses and a better merchandise mix to boost gross margins. Mitchell said Macy's has also improved delivery expenses due to reductions in packages per order and distance traveled.

More retail and commerce: How Home Depot Blends Art and Science of Customer Experience | How Wayfair's tech transformation aims to drive revenue while saving money | Connecting Experiences From Employees to Customers

Target

Target CEO Brian Cornell said, "consumers continue to rebalance their spending between goods and experiences and make tough choices in the face of persistent inflation."

Cornell added that Target is cautious about the near-term outlook but playing the long game by "investing in our stores, our supply chain, our team, our digital capabilities and our assortment."

Same-day services saw high single digit growth on a same-store basis. Cornell said customers are pressured by interest rates, loan payments and less savings. With discretionary income down, consumers are making tradeoffs and waiting for sales.

John Mulligan, Chief Operating Officer at Target, said the retailer has faced multiple challenges over the last 3.5 years including a lack of inventory, a demand boom, a downturn, inventory bloat and normalization.

Mulligan added that Target is cautious about its inventory position, but in a good place so far. Target has also benefited from lower supply chain costs. Mulligan said:

"We've seen improvements in metrics relating to backroom inventory accuracy and the percentage of new assortments set on time. In the digital channel, the percentage of orders picked and shipped on time and the average Drive-Up wait time have all improved from last year. Also, in support of the digital business, the percentage of items ordered but not found has declined from a year ago, meaning that we are fulfilling more items per order and canceling fewer, a key factor in guest satisfaction."

Target is also looking to improve its guest experience as measured by Net Promoter Scores in categories such as checkout, front-of-the store interactions, and digital services.

The Children's Place

Jane Elfers, CEO of The Children's Place, said the company has managed inventory well in the third quarter, but saw higher distribution and fulfilment costs as it pivots to more e-commerce.

The plan for The Children's Place is to hone its digital game given that's how Gen Z will buy when that generation has children.

Elfers said:

"Our digital channels are clearly where our current core millennial customer prefers to shop for her kids. And based on the data, digital is where our future Gen Z moms will overwhelmingly prefer to transact. Almost all new digital buyers will come from Gen Z. Gen Z digital buyers nationwide are expected to surge from 45 million today to over 61 million in 2027, only four short years away. The importance of the digitally native Gen Z demographic to our future business cannot be underestimated, and we remain laser-focused on ensuring that digital is at the core of everything we do."

The current generation of core customers for the retailer remains under pressure, but Elfers said The Children's Place is playing the long game with accelerated digital transformation and fleet optimization. The goal is to "operate the company with less resources, including less stores, less inventory, less people and less expense, allowing us to better service our customer online where she prefers to shop, resulting in what we believe will translate to more consistent and sustainable results over time," she added.

According to The Children's Place, the pivot to digital marketing has paid off including a focus on social media presence.

Gap Inc.

New Gap CEO Richard Dickson said the company is looking to strengthen its "operating platform" as it is retooling its core brands--Gap, Old Navy, Banana Republic and Athleta. He said:

"In some areas, we are in good shape, but we have more work to do. Our supply chain is a pillar of strength at Gap Inc., where our scale gives us unique cost leverage, but we need to accelerate innovation. Our financial strategy is driving early value, but we need to continue our focus on rigor and efficiency. In technology, we’ve made strategic investments and now it’s about optimizing those investments and driving adoption across the organization."

Dickson said the scale of Gap Inc. should be able to boost operating margins. Gap Inc. has four billion-dollar brands with 2,600 company operated stores and 1.4 billion annual visits to the company's websites. The company also has 58 million active customers.

Williams Sonoma

Williams Sonoma CEO Laura Alber's third quarter results delivered record operating margins of 17% as it benefited from customer experience improvements and lower supply chain costs.

The company projected fiscal year revenue to be down 10% to 12% but raised its operating margin outlook to 16% to 16.5%.

Alber acknowledged that consumer spending remains challenged, but a portfolio of brands has enabled the company to weather uncertainty even as same-store sales fall.

"Our in-house design capabilities and vertically integrated supply chain are also key in producing proprietary products at the best quality value relationship in the market," she said.

During the company's third quarter earnings call, Alber outlined the following strategies going into the holiday season.

• Introducing more new products at mid-tier and lower-tier price points without discounting.
• Sell through inventories with lower supply chain costs by reducing out of market and multiple shipments.
• Customer service improvements. Alber noted that Williams Sonoma's customer service metrics have returned to pre-pandemic levels as have on-time deliveries.
• Leverage investments in last-mile delivery to reduce customer accommodations, returns, damages and replacements.
• Continue to invest in digital experiences with content, tools for design projects and AI.

Dick's Sporting Goods

Dick's Sporting Goods Lauren Hobart said the company is targeting omnichannel athletes and giving them a good digital experience.

Hobart said:

"In combination with our stores, our digital experience remains an integral part of our success, and the investments we are making in technology are strengthening our athletes’ omnichannel experience and driving increased engagement. This quarter, we added 1.6 million new athletes and are further growing our base of omnichannel athletes. Omni channel athletes make up the majority of our sales and they spend more and shop with us more frequently than single channel athletes."

Hobart also said that the sporting goods retailer is investing in data science and personalization to create one-to-one relationships with athletes.

The company said that its consumer base is holding up well as they prioritize a healthy and active lifestyle.

Best Buy

Best Buy CEO Corrie Barry said "consumer demand has been even more uneven and difficult to predict" and the company lowered its fourth quarter revenue outlook. The technology retailer is focused on customer experiences, driving recurring revenue and offering new services.

"We continue to increase our paid membership base and now have 6.6 million members. This compares to 5.8 million at the start of the year," said Barry. My Best Buy Total is a $179.99 a year service that includes 24/7 Geek Squad Service, AppleCare Plus and two-years of product protection. My Best Buy Plus is a $49.99 a year tier that includes access to new products, two-day shipping, a 60-day return and exchange window and exclusive pricing.

Barry said the company is looking to drive interactions too.

"We have also seen growth in sales from customers who are getting help from our virtual sales associates. These interactions, which can be via phone, chat or our virtual store, drive much higher conversion rates and average order values than our general dot.com levels. This quarter, we had 140,000 customer interactions by a video chat with associates, specifically out of our virtual store locations."

In addition, the company continues to invest in its multichannel fulfillment operations. "As a reminder, while almost one-third of our domestic sales are online, 43% of those sales were picked up in one of our stores by customers in Q3. And most customers shop us in multiple channels," said Barry.

Best Buy is also bolstering its supply chain network to optimize the company's ship-from-store-hub and shipping locations to deliver with speed. About 62% of e-commerce small packages were delivered to customers from automated distribution centers. Those operations are supplemented with a delivery partnership with DoorDash.  

How much generative AI model choice is too much?

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

This post first appeared in the Constellation Insight newsletter, which features bespoke content weekly.

With enterprises still kicking the tires on large language models (LLMs), use cases and generative AI applications, vendors are big on providing choice, bring-your-own-models and the ability to mix and match foundations.

How much LLM choice is too much? We'll probably find out in the months ahead, but I’d argue we’re getting close. And the OpenAI kerfuffle is only going to invite more startups with LLMs looking to take share. 

Consider the following:

• Microsoft's Ignite conference featured Copilot Studio, a platform that will enable enterprises to use multiple models, create custom ones and extend Microsoft copilots. Microsoft will have a copilot for every experience. Meanwhile, Scott Guthrie, Executive Vice President, Cloud + AI Group at Microsoft, told analysts the company is committed to model choice beyond its OpenAI partnership. Guthrie said enterprises will ultimately be able to use one model to fine tune another and mix models with the same interaction context.

• Dell Technologies and Hugging Face announced a partnership aimed at on-premises generative AI. The aim is to give enterprises the ability to procure systems and tune open-source models more easily.
• Microsoft launched Azure Models as a Service with Stable Diffusion, Llama2 from Meta, Mistral and Jais, the world's largest Arabic language model. Command from Cohere will also be available.
• Google Cloud has its Model Garden, which features more than 100 models including foundational ones and task-specific ones.
• Amazon Web Services has Amazon Bedrock, which features models from AI21 Labs, Anthropic, Cohere, Meta, Stability AI and Amazon in a managed service. I'm no psychic but rest assured there will be a heavy dose of Bedrock at re:Invent 2023 later this month.
• OpenAI outlined plans to build out its ecosystem with custom GPTs and a marketplace to follow. 
• Software vendors are also enabling bring your own model arrangements, partnerships and development of domain specific models. ServiceNow's strategy plays both sides of the LLM equation--open LLMs as well as ServiceNow developed models. ServiceNow's Jon Sigler, senior vice president of Now Platform, said:

"We have a dual approach to models. We integrate general purpose models through a controller. We are also very focused on the smaller faster, more secure, less expensive models that do domain specific things. The combination of the large language models for general purpose and domain specific gives us the opportunity to do something that nobody else can do."

My hunch is that we'll reach some level of appropriate foundational model choice and then focus on small models aimed at specific use cases. Consider what Baker Hughes did with C3 AI and ESG materiality assessments. Specific model for a specific use case.

The current LLM landscape rhymes with that early big data buildout where the industry initially assumed they'd adopt open-source technology and then expand. The problem: For many enterprises, big data implementations never moved past the science experiment stage. There was no need for enterprises to run their own Hadoop/MapReduce clusters.

Will the average enterprise really know the difference between the latest GPT from OpenAI vs. Meta's Llama 2 vs. Anthropic's Claude vs. MosaicML? Probably not. That choice just brings complexity. The real market in LLMs will be domain-specific models that drive returns. Foundational models will be commoditized quickly and the difference between choosing between 50 models, 100 models and 1,000 models won't matter. A booming marketplace of custom models, however, will drive returns.

In the end, the massive selection of models will require an abstraction layer to make enterprise life easier.

OpenAI ousts Sam Altman as CEO over lying to board

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

Sam Altman is out as CEO of OpenAI.

In a surprise move, OpenAI's board of directors said Altman was out as CEO and board of directors. According to an OpenAI blog post, Mira Murati, the company's CTO, will be the interim CEO effective immediately. OpenAI President and Co-Founder Greg Brockman is also out. He outlined the Altman ouster on X and The Information reported that other of key executives resigning. 

"We too are still trying to figure out exactly what happened...The outpouring of support has been really nice; thank you, but please don’t spend any time being concerned. We will be fine. Greater things coming soon."

The board said that Altman's departure follows "a deliberative review process by the board, which concluded that he was not consistently candid in his communications with the board, hindering its ability to exercise its responsibilities."

OpenAI recently held its developer conference where it outlined plans for custom GPTs with Altman on stage. Microsoft also held its Ignite 2023 conference this week and CEO Satya Nadella called out OpenAI as a key partner.

In a statement, Nadella said:

"We have a long-term agreement with OpenAI with full access to everything we need to deliver on our innovation agenda and an exciting product roadmap; and remain committed to our partnership, and to Mira and the team. Together, we will continue to deliver the meaningful benefits of this technology to the world."

OpenAI launches GPTs as it courts developers, models for use cases | Software development becomes generative AI's flagship use case

Speaking at Ignite, Nadella said:

"They're just doing stunning breakthrough work. We are thrilled to be all-in on this partnership together. As OpenAI innovates we will bring that innovation to Azure AI."

While OpenAI said Murati is a good successor due to her unique skill set and "close engagement" with all aspects of the company, the AI firm will conduct a formal search for a CEO.

OpenAI was founded as a non-profit in 2015 and restructured in 2019 so it could raise capital. Microsoft has been the key partner of OpenAI and has popularized the use of ChatGPT for its copilot strategy. That said, Microsoft is also offering other models on Azure.

For enterprise buyers, the big question is whether the OpenAI management saga has any impact on Microsoft's copilot strategy. See:

• Microsoft Ignite 2023: Three big AI takeaways
• Microsoft launches Azure Models as a Service
• Microsoft launches AI chips, Copilot Studio at Ignite 2023
• Microsoft's full list of Ignite news and video.

In regulatory filings, Microsoft noted OpenAI in risk factors but didn't address management changes at its AI partner. In January, Microsoft announced the third phase of its partnership with OpenAI. 

Here's what Microsoft said in its annual report.

"We have a long-term partnership with OpenAI, a leading AI research and deployment company. We deploy OpenAI’s models across our consumer and enterprise products. As OpenAI’s exclusive cloud provider, Azure powers all of OpenAI's workloads. We have also increased our investments in the development and deployment of specialized supercomputing systems to accelerate OpenAI’s research."

In Microsoft's most recent quarterly SEC filing, the company said:

"We are building AI into many of our offerings, including our productivity services, and we are also making AI available for our customers to use in solutions that they build. This AI may be developed by Microsoft or others, including our strategic partner, OpenAI. We expect these elements of our business to grow. We envision a future in which AI operating in our devices, applications, and the cloud helps our customers be more productive in their work and personal lives. As with many innovations, AI presents risks and challenges that could affect its adoption, and therefore our business. AI algorithms or training methodologies may be flawed. Datasets may be overbroad, insufficient, or contain biased information. Content generated by AI systems may be offensive, illegal, or otherwise harmful. Ineffective or inadequate AI development or deployment practices by Microsoft or others could result in incidents that impair the acceptance of AI solutions or cause harm to individuals, customers, or society, or result in our products and services not working as intended. Human review of certain outputs may be required."

Alibaba hits brakes on cloud spin off plans

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

Read more

Alibaba said it won't spin off its cloud intelligence group citing US sanctions and an uncertain demand environment.

The company reported cloud revenue of $3.79 billion, up 2% from a year ago, for the quarter ended Sept. 30. EBITA for the cloud unit was $193 million. Alibaba recently launched a new proprietary large language model, built out its AI developer ecosystem and AI application development platform.

However, US expanded export controls are likely to hamper Alibaba's cloud unit and its ability to procure processors for AI inference and model training. Previously: Alibaba plans to spin off its Cloud Intelligence Group within 12 months

The company said in a statement:

"We believe that these new restrictions may materially and adversely affect Cloud Intelligence Group’s ability to offer products and services and to perform under existing contracts, thereby negatively affecting our results of operations and financial condition. These new restrictions may also affect our businesses more generally by limiting our ability to upgrade our technological capabilities."

Due to the uncertainty, Alibaba said it will hit the brakes on its cloud spin off. The company said it will focus on a "sustainable growth model for Cloud Intelligence Group under fluid circumstances."