Specialized Small AI Models: Reshaping the Future of AI Technology

By Carsten Krause, August 16, 2024

---

Is Generative AI’s Hype Fading? What Technology Leaders Need to Know

Generative AI has been like that flashy new gadget that everyone rushes to buy, only to realize later that it’s not quite the game-changer they thought it would be. Remember when 3D TVs were going to revolutionize how we watched television? Yeah, we all remember how that turned out (spoiler: it didn’t). Now, Gen AI might be finding itself in a similar spot, with the initial sizzle giving way to more than a few fizzle-worthy moments. But before you start thinking that AI is the next 3D TV, let’s unpack why this might not be the end, but rather a much-needed course correction.

The Rise of Specialized Small AI Models: Lean, Mean, and Focused Machines

The AI landscape is increasingly embracing a more “specialized” approach—think of it as trading in your Swiss Army knife for a precision tool. Sure, the Swiss Army knife can do a lot, but sometimes, you just need a good, old-fashioned screwdriver. This is where specialized small AI models come in. These models are designed to tackle specific tasks with laser-like focus, which is perfect when you don’t need the entire Swiss Army spread.

Consider the trend toward open-source AI models like Meta’s Llama 2 and Mistral AI’s Mixtral-8x7B. These models are giving the big-name proprietary models a run for their money. In fact, they’re often outperforming the big guns in certain areas—like the underdog team that comes out of nowhere to win the championship​ (Scribble Data). For companies worried about privacy, control, and cost (and who isn’t these days?), these specialized models offer a way to implement AI solutions without breaking the bank—or their data security protocols.

Moreover, the shift from cloud-based AI to on-device AI is like the difference between sharing a community swimming pool and installing your own private lap pool. Sure, the community pool has its perks, but sometimes you just want to do your laps without the splashes and distractions. Running AI on personal devices not only enhances privacy but also reduces latency, meaning your AI assistant doesn’t need to phone home every time you ask it a question​ (Scribble Data)​ (TechRepublic).

Navigating the Generative AI Correction: Avoiding the Potholes on the AI Highway

Generative AI’s initial trajectory was like a rocket shooting straight into the stratosphere—everyone wanted a piece of it, and it seemed unstoppable. But as with any rapid ascent, there’s bound to be some turbulence on the way down. Gartner predicts that by 2025, nearly 30% of Gen AI projects might end up abandoned in a ditch on the side of the AI highway. The reasons? Poor data quality, unclear business value, and costs that make your CFO break out in hives.

But let’s not write off Gen AI just yet. It’s more like a midlife crisis than a total breakdown. This correction is necessary to separate the hype from reality. In sectors like healthcare, retail, and eCommerce, Gen AI continues to transform operations. But here’s the kicker—success in these fields is increasingly tied to how well organizations can tailor AI to meet specific, actionable needs rather than chasing the latest trend like a kid running after an ice cream truck​ (Master of Code Global).

Sustainability is also becoming a driving force behind AI strategies. Imagine being a CIO who’s not only responsible for tech but also has to worry about your company’s carbon footprint—like being asked to juggle flaming torches while riding a unicycle. By 2027, Gartner predicts that 25% of CIOs will have their compensation linked to their sustainable technology impact​ (TechRepublic). That’s right, your next bonus might depend on how green your AI is.

The Rise and Potential Fall of Generative AI

The hype around Gen AI has been monumental, with organizations across the globe racing to integrate these capabilities into their operations. From intelligent chatbots to virtual assistants, the potential applications of Gen AI seemed limitless. Yet, as with any new technology, the initial excitement is often tempered by the harsh realities of implementation, scalability, and cost-effectiveness.

Gartner’s latest analysis places Gen AI past the peak of inflated expectations and heading towards the so-called “trough of disillusionment.” The firm predicts that by 2025, approximately 30% of current Gen AI projects will be abandoned after the proof-of-concept stage. The reasons are manifold: poor data quality, inadequate risk controls, unclear business value, and escalating costs are just some of the factors contributing to this trend.

To illustrate, consider the financial implications of implementing a Gen AI virtual assistant. According to Gartner, the initial rollout could cost between $5 million and $6.5 million, with ongoing annual costs ranging from $8,000 to $11,000 per user. Determining the ROI on such investments, particularly for ubiquitous applications like virtual assistants, is challenging. For example, while these tools might save employees time in searching for documents or composing emails, most organizations lack historical data to quantify such savings.

The Value of Traditional AI Technologies

As the hype around Gen AI cools, it’s worth remembering that traditional AI technologies—such as machine learning, deep learning, and predictive analytics—continue to deliver significant value. These technologies have proven their worth across a range of industries, from finance to healthcare, and remain critical components of many organizations’ AI strategies.

Chris Stephenson, Managing Director of Intelligent Automation, AI, and Digital Services at IT consulting firm alliantgroup, emphasizes the continued relevance of these technologies. “There are a lot of cool AI solutions that are cheaper than generative AI,” he notes. For many organizations, these traditional AI solutions are more than sufficient to meet their needs, often at a fraction of the cost of Gen AI.

Stephenson also points out that many AI solutions already exist within companies’ tech stacks, waiting to be leveraged. “When we do planning sessions with our clients, two-thirds of the solutions they need don’t necessarily fit the generative AI model,” he says. This suggests that CIOs should take a broader view of their AI strategy, considering all available options before committing to Gen AI.

Specialized AI Models: A Promising Alternative

As the shine of Gen AI starts to dull, many in the industry are beginning to shift their focus toward more specialized AI models. These models, designed to address specific business needs, offer a more targeted and often more effective approach to AI implementation.

“Specialized AIs, tailored to specific needs, offer more precise and effective solutions, delivering greater value and reliability for organizations,” says Hassan Uriostegui, CEO and co-founder of WakenAI. Unlike the broad-strokes approach of Gen AI, specialized models are fine-tuned to perform specific tasks, leading to more accurate outcomes and, ultimately, higher ROI.

Uriostegui’s perspective is shared by many in the industry, who see this shift as a necessary correction rather than a full-blown collapse. “The AI market is experiencing a correction, not a burst,” he adds. As the industry recalibrates, CIOs would do well to adjust their strategies accordingly, focusing on realistic applications of AI that align with their organizational goals.

The Role of Fine-Tuned Models

One of the key advantages of specialized AI models is their ability to be fine-tuned for specific use cases. This approach allows organizations to leverage existing AI models, making minor adjustments to better suit their needs rather than building new models from scratch.

Hamza Tahir, CTO and co-founder of ZenML, an open-source MLOps startup, advises CIOs to explore this approach before investing heavily in new Gen AI projects. “Fine-tuning existing AI models can often yield better results with less effort, allowing organizations to quickly realize the benefits of AI,” he says.

Fine-tuning not only reduces the time and cost associated with AI implementation but also minimizes risk. By starting with proven models and making incremental adjustments, organizations can more easily predict outcomes and ensure that their AI investments deliver the desired results.

Many organizations are looking at open source models to tailor to their needs, deploy securely and take advantage of a larger open source talent pool.

The Importance of Strategic Alignment

As CIOs navigate the evolving AI landscape, one of the most critical factors to consider is the alignment of AI initiatives with broader organizational goals. Too often, AI projects are undertaken in isolation, driven by the allure of cutting-edge technology rather than a clear business need.

CIOs should prioritize projects that address specific pain points or strategic priorities within their organizations. This approach ensures that AI investments are not only technically sound but also aligned with the company’s long-term vision.

For instance, while AI-driven chatbots and virtual assistants may be popular, they may not align with the strategic goals of every organization. Instead, CIOs should focus on AI applications that directly contribute to their organization’s mission, whether that’s improving customer service, optimizing supply chain operations, or enhancing product development.

Building Internal Capabilities

Another key consideration for CIOs is the development of internal AI capabilities. As AI becomes increasingly integral to business operations, organizations must build the skills and infrastructure necessary to support ongoing AI initiatives.

This includes fostering a culture of continuous learning and upskilling within the organization. By investing in training and development, CIOs can ensure that their teams have the expertise needed to effectively implement and manage AI projects.

Moreover, building internal capabilities allows organizations to be more agile in their AI initiatives. With the right skills in-house, commnity and partner support companies can quickly adapt to changing market conditions, experiment with new AI technologies, and scale successful projects across the organization.

Strategic Recommendations for CIOs and other technology leaders: The Blueprint for AI Success

1. Prioritize Specialized AI Models: When it comes to AI, think of specialized models as your company’s secret weapon. They’re the underdogs that pack a punch, delivering high ROI by solving specific problems rather than trying to be a jack-of-all-trades. It’s like hiring a top-tier chef for a fancy dinner instead of having your multi-talented cousin who’s okay at cooking, painting, and DJing all at once.
2. Leverage Open-Source Solutions: Open-source AI models are like the cool indie bands that everyone eventually realizes are just as good as the mainstream hits—maybe even better. They offer the flexibility to customize your AI to fit your organization’s needs without the restrictive licensing fees and rigid structures of proprietary models​ (Scribble Data).
3. Embrace On-Device AI: Moving AI from the cloud to personal devices is like upgrading from a dial-up connection to fiber optics—it’s faster, more secure, and tailored to your needs. This is especially critical for industries handling sensitive data, where privacy isn’t just a feature; it’s a necessity​ (TechRepublic).
4. Focus on Sustainability: As environmental concerns become a boardroom priority, your AI strategy needs to align with these goals. Imagine having a tech solution that’s not only smart but also green. It’s like driving a Tesla instead of a gas-guzzler—not only do you get to feel good about your carbon footprint, but you also set a powerful example for others in your industry​ (Master of Code Global).
5. Avoid the Gen AI Trap: It’s easy to get caught up in the hype, but as with all trends, what goes up must come down. Don’t throw all your resources into a single Gen AI basket. Instead, diversify your AI portfolio, integrating traditional AI models and specialized solutions to create a well-rounded strategy that’s resilient to market fluctuations.

The CDO TIMES Bottom Line

Generative AI is like that new gadget you thought you couldn’t live without—until you realize it’s collecting dust on the shelf. As the initial hype begins to fade, CIOs, CDOs, and CXOs need to look beyond the glitz and glamour and focus on what really drives business value. Specialized small AI models offer a practical, cost-effective alternative that aligns with both business needs and sustainability goals.

So, don’t let the Gen AI craze lead you astray. Stay grounded, keep your strategies diversified, and always remember that in the world of AI, sometimes less really is more.

For more insights on navigating the evolving AI landscape, subscribe to The CDO TIMES. Visit https://www.cdotimes.com/sign-up/ to become a paid subscriber today.

Love this article? Embrace the full potential and become an esteemed full access member, experiencing the exhilaration of unlimited access to captivating articles, exclusive non-public content, empowering hands-on guides, and transformative training material. Unleash your true potential today!

Order the AI + HI = ECI book by Carsten Krause today! at cdotimes.com/book

Subscribe on LinkedIn: Digital Insider

Become a paid subscriber for unlimited access, exclusive content, no ads: CDO TIMES

Do You Need Help?

Consider bringing on a fractional CIO, CISO, CDO or CAIO from CDO TIMES Leadership as a Service. The expertise of CDO TIMES becomes indispensable for organizations striving to stay ahead in the digital transformation journey. Here are some compelling reasons to engage their experts:

1. Deep Expertise: CDO TIMES has a team of experts with deep expertise in the field of Cybersecurity, Digital, Data and AI and its integration into business processes. This knowledge ensures that your organization can leverage digital and AI in the most optimal and innovative ways.
2. Strategic Insight: Not only can the CDO TIMES team help develop a Digital & AI strategy, but they can also provide insights into how this strategy fits into your overall business model and objectives. They understand that every business is unique, and so should be its Digital & AI strategy.
3. Future-Proofing: With CDO TIMES, organizations can ensure they are future-proofed against rapid technological changes. Our experts stay abreast of the latest AI, Data and digital advancements and can guide your organization to adapt and evolve as the technology does.
4. Risk Management: Implementing a Digital & AI strategy is not without its risks. The CDO TIMES can help identify potential pitfalls and develop mitigation strategies, helping you avoid costly mistakes and ensuring a smooth transition with fractional CISO services.
5. Competitive Advantage: Finally, by hiring CDO TIMES experts, you are investing in a competitive advantage. Their expertise can help you speed up your innovation processes, bring products to market faster, and stay ahead of your competitors.

By employing the expertise of CDO TIMES, organizations can navigate the complexities of digital innovation with greater confidence and foresight, setting themselves up for success in the rapidly evolving digital economy. The future is digital, and with CDO TIMES, you’ll be well-equipped to lead in this new frontier.

Do you need help with your digital transformation initiatives? We provide fractional CAIO, CDO, CISO and CIO services, do a Preliminary ECI and Tech Navigator Assessment and we will help you drive results and deliver winning digital and AI strategies for you!

Subscribe now for free and never miss out on digital insights delivered right to your inbox!