With all the hype in the past several years around AI (or more specifically GenAI), I'm not afraid to say – I'm an AI skeptic.
It doesn't mean that I don't believe that some day AI may have a huge impact on human beings' lives, but at the moment, all I can see is irrational hype.
In my background, I came from infra-security; I am not a developer, nor do I consider myself an AI expert. I am a cloud architect, meaning I'm looking at proposed architectures, seeing how they suit business requirements, and whether they are deployed in a secure, resilient, and perhaps cost-effective way.
I don't see value in adding AI to every design, just for the sake of saying "our application now includes AI".
I've been watching the industry since 2023 go nuts. Suddenly, everyone is eager to add AI capabilities, chasing some unexplained FOMO before the machines replace our jobs.
I'm not against the use of AI. As a matter of fact, I've been using Grammarly for many years (since, for most of us, English is not our first language).
In the past several years, I've been using chatbots such as ChatGPT, Perplexity, and recently Gemini daily, asking questions about various topics and aspects of my life.
From asking the bot to provide me an answer about a specific character in a favorite TV show, to "how do I resolve an alert shown on my car's dashboard," and up to "summarize this blog post for my newsletter".
It's great that I can ask Gemini to create me a LinkedIn post based on an article I just read, add some emojis and hashtags, and at the end create me a cover image for the post.
For a probabilistic system, this is great. I am expecting the system to be creative and produce me attractive results, sometimes even funny images.
For a home consumer, this is great, but far from been ground breaking technology.
I truly believe that the "big money" will come from enterprises paying a lot of money for AI-based solutions, once the industry can actually make something good from a non-deterministic system.
From an enterprise point of view, if I'm querying a database, I'm expecting the same result every time.
Think about life-changing revolutions such as finding a cure for cancer, or for the next pandemic.
How can humanity truly trust technology to produce reliable and sustainable results using the current state of AI capabilities?
Although as humans we tend to see AI as something amazing, at the end of the day, these are simply language models, trained on human knowledge (supervised and unsupervised learning). This is far from being an artificial intelligence with actual self-learning capabilities or feelings, knowing the difference between right and wrong.
All it can generate is something based on pre-trained material (which some large LLM vendors have already proven that their models were trained based on stolen intellectual property from authors such as myself).
It is true that, from time to time, we hear customers' stories about the use of AI capabilities deployed in production, but as we currently know, most AI projects fail due to various reasons (from lack of architecture design, knowledge gap, cost management, etc.)
At the moment, there is a huge hysteria in almost every industry, caused by journalists who fail to ask questions such as what we actually get from AI, or does it actually produce real value for humans?
Tech analysts are running from one vendor conference to another, spreading rumors of how vendor X released the best AI model, replacing a competitor's model Y in being the latest and greatest (but mostly the most expensive for the rest of us to be able to consume daily).
CEOs have a FOMO that they will fail to jump on the AI train, which forces them to push AI into every corner of their organization, regardless of whether it makes sense or not. Just to make sure their board (and, if they are startups, their VCs) are happy.
Billions of dollars are spent every year trying to catch up with the AI hype, throwing money at expensive NVIDIA GPUs (which are getting more and more expensive every quarter) and expensive closed-source LLMs.
Employees are forced to use AI, even when it doesn't really make sense, and alternatives such as automation capabilities make more sense.
One area where most of the tech industry believes is the best use case for AI is development.
At first, it used to make sense that if AI coding assistance can recommend lines of code, perhaps we can throw coding tasks to AI coding assistance, cut the size of current development teams, and perhaps we don't even need junior developers, since their abilities are more or less coding assistance.
The truth is far from the current belief.
What managers fail to understand is that regardless of the amount of code that AI was able to generate, people remain the full owners of systems.
AI may be able to deploy applications, even at scale, but when things go wrong (and everything eventually fails), you still need a skilled developer to be able to understand the written code, debug it, and understand what was written and for what purpose.
Maintaining large-scale production systems is the real challenge.
Taking decades of legacy code (from ancient Mainframe systems to legacy Java, Python, and .NET applications) that should be modernized, decoupled, reviewed for application security vulnerabilities, and unfixed bugs, based on hard-to-maintain monoliths – this is a huge challenge that organizations haven't been able to resolve so far (though many of the legacy systems still produce business value…)
Nobody ever said that the more code humans generate (or recently code generated by AI) is good for anything. It's just a ton of AI slop that brings no real and sustainable value for humans.
Don't get me wrong – I am still reading about AI, trying to understand what someone like me, with a non-developer background, can actually do with this technology, trying not to lose my current job, hoping to remain relevant (even after 25+ years in the tech industry).
As a cloud architect, studying public cloud services from AWS, Azure, and GCP for almost a decade, I still believe that the public cloud is the best place for most enterprise applications (although the current price for consuming services from the hyper-scale cloud providers might make me thinking that organizations should push the hyper-scale cloud provided to dramatically lower their service cost or lose the bottle for neo-clouds and co-location providers).
I don't see GenAI as a technology revolution. It has a lot of potential, but it is still an evolution of existing software technology.
I don't believe Skynet is coming to get us anything in the near future, nor do I see AI taking humans' jobs.
It may replace some human repetitive tasks, but it cannot replace humans with their creativity, passion, and, let's not forget – human ownership.
We will still need human-in-the-loop for every critical decision, similarly to the saying "the fact that you're using Waze or Google Maps doesn't mean that you can turn off your brain".
Don't ignore AI – keep learning, and put things in perspective.
About the Author
Eyal Estrin is a cloud and information security architect, with more than 25 years in the industry.
He is the author of Cloud Security Handbook and Security for Cloud Native Applications.
The views expressed are his own.
Top comments (1)
I think the strongest version of the skeptical case is budget discipline. Some tasks deserve a model, some deserve boring automation, and many deserve no new tool at all. The governance question is whether teams are allowed to say that before the GPU invoice arrives.