Here we are, right at the beginning of what some are calling 'The Dawn of AI'. It makes you wonder, doesn't it? Are we going to look back and smile at these times, or will we wish we could hit the undo button? This moment is undeniably significant. We're at the start of a big change in how we use technology and connect with each other.
But let’s pause and think about what has really changed. “AI is suddenly everywhere,” you might remark. That’s true now more than ever. But it’s been like this before; it was just less obvious. AI is curating your social feed, suggesting items for your shopping cart, deciding what ads you see, and what search results are the most relevant. AI’s reach has been extensive for quite a while. Yet, the landscape of AI was once the exclusive domain of brilliant scientists and dedicated engineers, whispering about neural networks and hidden layers over cups of coffee. Now that generative AI is here, everyone’s invited to the party. This is where the main difference is: before, you had to obtain special knowledge and spend a lot of time training an AI model to solve a specific problem, and now powerful AI tools that can solve a bunch of different problems are at everyone's fingertips. This isn't just a game changer, it's a whole new game being played in ways we’re just now beginning to grasp.
In the current environment where everything is “the next big revolutionary AI tool," it's surprisingly easy to lose sight of the essence of this transformation. For all of us, separating the practical from the fantastical is crucial. This blog post aims to untangle the AI hype, offering a clear-eyed view of AI's current state, its genuine value, and how we can navigate this evolving landscape with wisdom and pragmatism.
Disclaimer: For simplicity, when we talk about generative AI technologies such as ChatGPT, Claude, Gemini, Copilot, and others, we’ll use the term "artificial intelligence" or "AI." While "traditional AI" still plays a role in research and development, our discussion will primarily center on generative AI, which is currently capturing the most attention.
Where we are now
We can say without doubt that the current wave of AI advancements started with ChatGPT. Earlier versions laid the groundwork, but ChatGPT's ability to chat really caught everyone's eye. It wasn’t the technical improvements in precision, speed, or quality, but the increase in accessibility and bridging the gap between technology and everyday users that made the difference and marked a leap in text-based AI tools and applications.
Going Beyond Text
The revolution, however, didn't stop at text. Within a remarkably brief span, it branched into other domains like image generation, video creation, music composition, and more. This expansion, coupled with significant upticks in usage and substantial investments pouring into the field, has attracted some of the world's brightest minds to explore and push the boundaries of what AI can achieve.
Whether we are currently at the peak of AI enthusiasm or it's on the way to climb even higher remains to be seen. Yet, one thing is sure: we’re seeing a fair share of groundbreaking claims, futuristic promises and an abundance of AI-powered gadgets and services, ranging from smart fridges to automated dog feeders and intelligent shoes. Amidst all this, distinguishing truly transformative advancements from mere novelties presents a real challenge. Much like the cryptocurrency boom not too long ago, the current explosion of interest in AI can serve as a reminder of the critical need for practicality and necessity in the realm of technological progress.
AI’s Limitations and Promises
Amid all the excitement and the tendency for some to oversell its capabilities, identifying the true limitations of artificial intelligence can be challenging. Even though AI has come a long way, it's not perfect. There are still a bunch of areas where it's trying to catch up. We will focus on the areas where it performs poorly compared to human abilities and skip other limitations such as bias, resource intensity, etc.
Artificial intelligence often struggles with understanding the full context of a conversation or situation, a task that humans usually handle without much effort. Similarly, comprehending the ethical implications of their actions or decisions remains a hurdle. Even with safeguards in place, these measures aren't always bulletproof and can be sidestepped. It's also common for these systems to produce errors or "hallucinations"—instances where the AI confidently presents incorrect or nonsensical information in an intricate and believable way. A critical limitation that ties this all together is the inability to assess the quality of its own responses, lacking the self-awareness to recognize inaccuracies or inappropriateness in the information it provides.
While this technology is a powerful tool in skilled hands, expecting it to produce flawless results without human supervision is currently unrealistic. These limitations remind us where AI is right now, but at the same time, they show a roadmap for future advancements towards even more sophisticated AI systems, which are bound to happen.
Where we are going
With the latest buzz around OpenAI's Sora and similar breakthroughs, it's easy to turn to questions like "What's coming up next?" or "Where is all this heading?" Questions like, "Will we still be employed in 5 years, or will AI have taken over most tasks?”, “How will the education system evolve?” or “Is there such a thing as AI becoming too clever for our own good?"
Honestly, no one has a clear picture of our destination or the best route to reach it safely and optimistically. Chatting about AI's future and how society will adjust could be a great way to spark lively discussions over a cold one. And if I let generative AI wrap up this thought, it might say something like:
"In this era of technological evolution, playing an active role is crucial. Encouraging ongoing learning, critical thinking, and ethical mindfulness paves the way for an AI future that is inclusive, responsible, and genuinely groundbreaking."
But let's be real—I'm not entirely sold on that. It's unlikely that you or I will play a monumental role in shaping AI's future. Speculating about what’s to come can easily become a distraction, a way to lose time and obscure the immediate and actionable steps that we should focus on. Therefore, I propose we shift some of our attention to a more practical question:
"How can I prepare for AI's next advancements, no matter the direction?"
As we move forward, understanding what we should do will become clearer, but regardless of what that will be, the initial step is invariably the same: becoming acquainted with and skilled in AI.
Practical Tips
So what can we do to get proficient in AI, especially while we are being bombarded with new things from every direction? Here are some practical tips that can help you become more knowledgeable in the age of AI.
Keep Learning and Growing Beyond AI
The first tip is not an AI tip. Continue learning and building your non-AI expertise. Even with AI everywhere, continuing to learn is key to staying ahead. If you are discouraged because AI can do more with each passing day, you may fall into the trap of over-reliance on AI. This will dull your competitive edge. Focusing too much on AI will make you a master at crafting prompts, but are you comfortable with that being the pinnacle of your career? Remember, at its core, AI remains a tool. A powerful one, yes, but it’s human insight and judgment that amplify its utility. Being able to critically assess the effectiveness and appropriateness of AI-generated content will depend on persistent learning and hands-on experience in your domain of choice.
Stay Reasonably Updated
It's easy to get lost in all the tech news out there. Find a few good resources (check out the suggestions at the end of this article) and stick with them. Combine sources that aggregate news (they will likely all report on the same things anyway) and a few more niche sources that are relevant for your interests. And then, don't just passively absorb information. Dive into the tools and techniques you learn about. Experiment with them, test them out, and form your own opinions. Be ready to sift through some noise to find the gems.
Breadth and Depth
Trying out new tools is important, but a lot of tools are variations of each other. While some are open-source, others handle images better, and some can provide sources or support multiple languages. But strip them down to their core, and you will find a set of the same functionalities. The real trick lies in finding the right balance between the exploration of new tools and diving deep into a select few that help you the most. Aim to identify the minimal set of tools that fulfills your requirements, as each additional tool will occupy valuable mental real estate.
Persevere
When we started using generative AI, we were impressed. It is levels better than anything we have seen before. Interacting with it feels as natural and effortless as chatting with a friend. The level of understanding is truly amazing, even with rushed, error-ridden prompts. Yet, it didn’t take long to realize that AI, despite its brilliance, doesn’t quite match up to its hype. A quick, simple prompt won't instantly create a hit book or a job-winning application. Just as we once learned to refine our search queries for better Google results, mastering the art of crafting effective prompts is a skill that will require effort.
There is a threshold in the quality of prompting that you must reach. Below it, you will see minimal improvements in the AI’s responses, even though you have spent time and improved your prompt. The reason for this is that AI is quite forgiving and will produce decent responses to less than perfect prompts. But to unlock its full potential, your prompts need to hit a certain level of quality. Reaching this point can be challenging, and it’s likely that you will be discouraged along the way. You might even say something I’ve heard a thousand times: “It’s not there yet”. And by the time it's “there,” you'll have missed out on valuable learning and growth opportunities. Remember, becoming proficient with AI is much like it is with any other skill, it requires patience, practice, and perseverance.
Think Broadly
It’s essential to not view AI as a final touch but as a tool that can be used throughout problem-solving. When you are tackling a challenge, there are multiple steps. It usually starts with data, which you need to organize and analyze. Then you draw insights and conclusions on which you base your decisions. The last step is to transform these decisions into a process. AI can be integrated at every step, helping you understand your data better and, in turn, make better decisions. Usually, it can also help you automate a part of the final process.
Starting with AI early can also help you spot any mistakes in your plan or notice things you might have missed. Think of AI as a helpful tool that makes it easier to understand your problem (before you even start working on it) by looking at it from different sides. This way, you're not just finding solutions faster, but you're also getting better at analyzing problems, which makes you better at solving them.
As we look ahead, the future seems both exciting and daunting. We can expect further integration of AI into all areas of our lives, beyond what seems reasonable or doable today. Looking even further ahead, who knows what will happen. What we do know is that this is a significant shift in how we work and live. All of us will be affected, so we need to be as ready as possible. Ignoring artificial intelligence is no longer an option.
