The State of Generative AI: Current and Future Outlook
The journey from the fictional AI entity H.A.L. in “2001: A Space Odyssey” to the emergence of Generative AI tools like ChatGPT and BARD marks a significant evolution in AI technology. It wasn’t until 2022 and 2023 that Generative AI, reminiscent of H.A.L. in its capabilities, gained widespread prominence, demonstrating the potential of what AI could accomplish with simple prompts.
With AI now firmly embedded in the public consciousness, the landscape seems poised at an inflection point. A Gartner Poll reveals a substantial shift, indicating that 55% of organizations are actively engaged in “piloting or partnering” initiatives with Generative AI—a notable 45% increase since the onset of 2023. According to a Gartner analyst, organizations are not merely discussing Generative AI; they’re actively dedicating time, financial investments, and resources to propel its development and drive tangible business outcomes.
The realm of AI transcends mere financial investments. With each iteration, Generative AI experiences exponential growth and compounding advancements. To understand the trajectory of these trends, let’s examine the current state of Generative AI and attempt to forecast where this technology is leading us.
The State of Generative AI: Where We Currently Stand
Even a casual phone or desktop user can’t ignore the ever-burgeoning technology of Generative AI. It even shows up in our email bodies, offering to help us craft communications. Let’s examine where Generative AI has already made inroads:
Generative Text Tools are The Most Common Forms of AI
According to a survey by Writer.com, ChatGPT is the most-used AI tool, with 47% of companies already employing it. That can cover a broad range of uses. ChatGPT can be a voice assistant for bouncing brainstorms off of. It can provide outlines for articles. It can even come up with rough drafts for emails, web pages, or ebooks. For people who need to fill web content but never have any idea where to start, it’s a game-changing piece of technology.
LLMs Have Changed How We Perceive AI
All of the above is possible through Large Language Models, or LLMs. It’s not that modern AI “thinks” in the way “2001: A Space Odyssey” was suggesting that H.A.L. might think. Instead, Large Language Models have adapted human language patterns and made them visible to the general public.
These LLMs analyze existing text and predict future text. It sounds innocuous, but the ramifications are profound. LLMs can translate, summarize, analyze, and even build new content as long as the LLMs have large enough data sets to work with. And as the digital world expands and LLMs grow more sophisticated, those data sets only get larger.
The result: Generative text AI sounds more “credible” and human to us than ever.
The Problems with Current Iterations of Generative AI
Generative AI doesn’t think. Generative AI creates new data or content that imitates human-like creativity and behavior. IT is designed to generate content, based on patterns and data it has been trained on. Generative AI sometimes shares flaws with human language and content based upon the data upon which it was trained. If a general reader without much exposure to a topic asks Generative AI to summarize a book, that reader will have no idea whether the AI is being accurate—or simply making the wrong verbal predictions based on text patterns.
This has already caused some issues in the legal world. Some attorneys have used ChatGPT to build arguments, only to find that Generative AI was simply making up case precedents, which would be invalid in a court of law. If LLM is going to represent the future of AI as well as its present, it’s going to require a greater degree of accuracy—maybe even exceeding human-quality research.
There are generally four pillars through which LLM can analyze content:
- Translation: One of the most reliable methods for using Generative AI, translation refers to moving from one language to another—even if that translation is explaining a line of code to a programmer. Coding languages are just as important here as spoken languages; Generative AI’s use of each brings greater accessibility to the average user.
- Summary: Generative AI’s ability to take vast swaths of information and express it succinctly can essentially “read” information for the user in advance. The question is: how accurate is the summary?
- Classification: When looking for a summary or a translation, Generative AI also needs to accurately classify content into specific categories. This helps form an accurate picture for the reader.
- Generation: Ultimately, the quality of Generative AI comes from its output. Does it make sense to human readers? Can humans use it?
The Future of Generative AI
Overall, the future of generative AI holds promise for transformative applications across numerous industries, with models becoming more adept, specialized, and integrated into various aspects of our daily lives. However, ethical considerations, continued research, and responsible deployment will be crucial in shaping a positive trajectory for these technologies.
Given the current trends, it’s fair to say that as Generative AI and the LLMs that fuel its progress advance, the quality of the AI will improve. The Generative AI tools we use will get stronger, more accurate, and more advanced.
But what about the uses for Generative AI which are still up in the air? What does that future look like? Here are a few possibilities for the directions Generative AI may take us.
Automation of Knowledge Work
The advancement and potential of Generative AI in automating knowledge work represent a significant shift in the way tasks and responsibilities are approached across various fields. according to a report from McKinsey, the imminent possibility of automating tasks in knowledge-intensive sectors, offering insights into how AI, particularly Generative AI, can transform and augment these domains.
- Creativity in Knowledge Work: Generative AI’s improvements can contribute to the creative aspects of highly cognitive fields, enabling the automation of tasks that traditionally required a human touch. This could potentially allow professionals in education, law, technology, the arts, and other sectors to delegate certain aspects of their work to AI systems.
- Democratization of Expertise: The traditional monopoly of certain skill sets could change with Generative AI. Instead of solely relying on highly skilled individuals, AI could democratize access to high-quality solutions. For instance, instead of needing a highly skilled coder, AI systems could provide coding solutions, democratizing access to coding capabilities.
- Impact on Regulated Professions: Fields like law and medicine, where expertise and experience are essential, may see a significant impact from Generative AI. While these sectors have strict regulations and requirements, AI’s advancements might challenge the traditional models by automating certain aspects of their tasks, potentially disrupting how expertise is leveraged and delivered.
- Integration and Ethical Considerations: As Generative AI becomes more involved in knowledge work, there will be ethical considerations and integration challenges. Ensuring that AI systems adhere to ethical standards, especially in regulated professions like law and medicine, will be critical.
- Augmentation, Not Replacement: The future is likely to see AI complementing human capabilities rather than completely replacing skilled professionals. Instead of taking over entire roles, AI may assist professionals by automating routine tasks, freeing them to focus on higher-value and complex aspects of their work.
The rise of Generative AI in automating knowledge work signals a profound shift in how various industries operate. While this advancement promises efficiency and innovation, it also raises crucial ethical and societal questions that need to be addressed responsibly as these technologies evolve.
Disproportionate Benefits to Some Industries Over Others
Generative AI also can potentially change the lay of the land across multiple industries. For example, Generative AI’s impact on software engineering can be incredible. Generative AI can help skilled professionals generate lines and pages of code at a moment’s notice—reducing the need for manual work and extending the productive capacity of each developer.
On the other hand, AI may not have quite the same impact on an industry like consumer packaged goods. Automated industrial processes don’t rely on Generative AI to improve processes—not without a similar improvement in the industrial hardware that drives these sectors.
AI could also vastly increase the quality of cybersecurity in financial services—helping them detect fraud and ID theft, for example. AI wouldn’t have such a drastic impact on other areas of security.
What these impacts are—and how they’ll resonate throughout the economy—are just beginning to develop. But one thing is certain, Generative AI’s rollout to general public use is going to usher in a new era of digital interaction. When AI can translate, speak, paint, and write, it has the potential to step in for a number of professional purposes. For now, it makes us more efficient but harnessing that efficiency can be challenging. Reach out to the team at comport to help you understand where AI can accelerate processes and improve your organization’s technology solutions.
