AI and its future: beyond the data-driven era
Contrary to current trend towards ever-larger models, the future may belong to Small Language Models (SLMs)
Artificial intelligence is the science of making machines do things that would require intelligence if done by humans — John McCarthy, who coined the term ‘artificial intelligence’ and is considered father of AI, said in 1955
Artificial Intelligence is the buzzword that’s resonating across boardrooms, classrooms, and coffee shops these days. It is everywhere. From chatbots handling customer service to algorithms curating social media feeds, AI has become the in-thing of our time. Yet despite the widespread adoption and breathless headlines, we’re still in the earliest stages of what AI can become.
The current reality: data-driven intelligence
Today’s AI systems, impressive as they may seem, operate on a fundamental principle: processing vast amounts of data to recognize patterns and generate responses. These Large Language Models (LLMs) can write poetry, code software, and answer complex questions, but they’re essentially sophisticated pattern-matching engines drawing from enormous datasets.
Frankly speaking, what we’re experiencing now is just the tip of the iceberg and we’re still in the fetal stage of artificial intelligence evolution.
However, the current data-driven approach has undeniably been disruptive. Industries from healthcare to finance have scrambled to integrate AI tools, leading to the ubiquitous presence of ‘AI-powered’ solutions. However, calling these systems true artificial intelligence may be premature - they lack the fundamental cognitive abilities that define genuine intelligence.
The next frontier: Artificial General Intelligence
The next phase in AI evolution promises something far more sophisticated: Artificial General Intelligence (AGI). Unlike current systems that excel in narrow domains, AGI will possess the ability to understand, learn, and apply intelligence across a broad range of tasks - much like human cognitive flexibility.
The key differentiator lies in cognition. Where today’s AI relies on statistical analysis of training data, AGI systems will develop the capacity for genuine reasoning and decision-making. This cognitive leap represents a fundamental shift from pattern recognition to actual thinking.
AGI won’t just process information faster or access more data - it will understand context, make inferences, and adapt to entirely new situations without requiring additional training. This represents a qualitative, not just quantitative, advancement in machine intelligence.
The ultimate goal: Absolute Intelligence
Beyond AGI lies an even more ambitious target: Absolute Intelligence. This final phase envisions AI systems with fully developed cognitive abilities - machines that can think, reason, and make decisions with the same depth and nuance as human consciousness, potentially surpassing human intellectual capabilities.
Absolute Intelligence would mark the point where artificial systems achieve genuine understanding rather than sophisticated mimicry. These systems would possess creativity, intuition, and the ability to grapple with abstract concepts in ways that current AI cannot.
Small Language Models: The Future Architecture
Contrary to the current trend towards ever-larger models, the future may belong to Small Language Models (SLMs). These more efficient, specialized systems could prove more practical and powerful than their data-hungry predecessors.
Small Language Models offer several advantages over massive LLMs: reduced computational requirements, faster processing, greater customization for specific tasks, and the ability to run locally rather than requiring cloud infrastructure. As AI becomes more integrated into daily life, these characteristics will prove increasingly valuable.
The shift toward SLMs reflects a maturation of the field - moving from brute-force approaches that require enormous resources toward elegant, efficient solutions that deliver superior performance with less overhead.
The way forward
Rather than dwelling on dystopian scenarios, the AI revolution presents an opportunity to thoughtfully shape the next decade of technological development. The progression from today’s data-driven systems through AGI to Absolute Intelligence won’t happen overnight.
However, the key lies in recognizing that we’re not approaching an endpoint but rather embarking on a carefully planned journey. Each phase of AI development builds upon the previous one, creating opportunities to refine our approach, establish ethical frameworks, and ensure that artificial intelligence helps humans. As we stand at this inflection point, the question isn’t whether AI will transform our world - it’s how we’ll guide that transformation. The next ten years will determine whether we harness these emerging capabilities to solve pressing global challenges, enhance human potential, and create a more prosperous future for all.
The age of true artificial intelligence is still ahead of us. What we’re witnessing today is merely the opening chapter of a much larger story - one that we have the power to write thoughtfully and purposefully. All said and done, the world needs a responsible AI that can enhance our quality of life in all spheres and spaces. That’s the bottom line.
(Krishna Kumar is a technology explorer & strategist based in Austin, Texas in the US. Rakshitha Reddy is AI developer based in Atlanta, US)
