Although AI has existed for several years, the recent advancements of large language models (LLMs) like GPT-4 have made these tools more accessible and prolific in the business world. Following the popularity of models like Dall-E and ChatGPT with the public, individuals and companies are increasingly recognizing the various applications of generative AI — and making it happen. A new dawn seems to have broken, offering a radical reimagining of how we get things done.
For the enterprise, adopting gen AI lies in pinpointing the crucial areas that align most closely with your core operations. When the iPhone came out, enterprises ran to build mobile apps without considering user adoption and experience. Leaders should learn their lesson here and be thoughtful about focusing on the key pain points and needs specific to your business. This ensures a more targeted and effective implementation of Gen AI. This selective approach will allow you to leverage the technology where it can make the most substantial impact.
Hotspots of gen AI adoption in today’s tech landscape
Within the enterprise, the hotspots of gen AI adoption are especially prevalent in coding assistance, content generation, and information retrieval. For example, AI co-pilots have become indispensable tools for developers, offering real-time code suggestions and enhancing the efficiency of software development. These co-pilots can understand and anticipate programmers’ needs, significantly accelerating the coding process. AI co-pilots can also perform the following:
- Extract important data and hidden insights from
- Structured and unstructured documents, including contracts, invoices, financial statements, resumes, knowledge repositories, and emails.
- Search and converse with these documents using natural language to quickly and easily find what users need.
- Summarize the contents of these documents and match documents using criteria defined in natural language.
Content generation tools powered by gen AI are transforming the way we create written content. They are used from marketing copy and creative writing to automated news articles and reports, enabling users to produce high-quality text quickly.
Additionally, gen AI has greatly improved information retrieval through natural language querying. Although it’s early, this leap looks like it will enable users to interact with data and databases without having to learn SQL. The tremendous value unlocked underscores the profound impact of gen AI on enhancing productivity and expanding the capabilities of technology across the enterprise.
Advancing gen AI beyond retrieval into the realm of reasoning
Clearly gen AI today has massive potential, but it’s still relegated to information retrieval. I believe the next big shift will come from agentic AI – a type of AI that can understand context and take actions based on what it knows. It’s not just about providing information. The real jump will come from products that can take information, understand what additional context is needed, and know how to route to the correct action.
We’re already seeing early versions of this agentic AI using LLMs in academia. Models like React are now proving that you can replicate human-level task solving while controlling hallucinations and showing step-by-step reasoning.
These developments are not confined to academia alone; early-stage startups are actively embracing the challenge of advancing agentic AI. Notably, the legal tech sector is witnessing a compelling array of companies pioneering AI systems capable of multistep reasoning, aimed at unraveling complex legal concepts.
This represents a significant leap forward from the current state of large language models, marking a pivotal transition from predictive proficiency to the realm of intelligent reasoning. This shift signifies a crucial step forward, from the proficiency of prediction to the intelligence of reasoning, ultimately redefining how AI can be used.
Redefining gen AI software business models
AI will change the way businesses are able to price. Since AI can generate outcomes, companies can, in essence, sell AI based on value capture. When the cloud emerged, software saw a huge transformation. Companies like Salesforce ushered a transition from traditional software licensing to the Software-as-a-Service (SaaS) model. In the past, purchasing software meant owning a piece of software that you had to manage and upgrade independently.
Salesforce and similar SaaS providers offered a subscription-based model where companies no longer owned the software but relied on the service provider to handle deployment, management, and upgrades. This was a dramatic shift as it redirected software expenditure from capital expenditures (CAPEX) to operating expenditures (OPEX). This unlocked new budgeting possibilities, as costs could be spread and amortized over extended periods, fostering greater flexibility and financial efficiency.
Our bold prediction is that AI will open up new spend because outcomes can be tied directly to the software, increasing the leverage and pricing potential that software vendors can command. Furthermore, with these tools, business models that have been eschewed by software companies in recent years, like services, can make a comeback.
Historically, customization was associated with expensive and labor-intensive service work. Software companies were generally averse to providing custom solutions because it impacted margins and scalability. However, gen AI can potentially allow for the creation of highly customized solutions at a fraction of the cost of traditional service-based approaches.
As mentioned earlier, we are already seeing many companies leverage gen AI in their engineering organizations to generate code. As AI tools continue to advance, the potential for achieving the same level of customization and personalization that services once offered is becoming increasingly feasible. Ultimately, gen AI software models are poised to empower businesses like never before, offering tailor-made solutions that align more closely with their unique objectives.
Be First to Comment