We’ve sort of considered artificial intelligence to be a buzzword. It’s a very loud buzzword, and while it has permeated into our world faster than most emerging technologies, there is a ton of controversy and an even more general misunderstanding of just what AI can do, and what the future might look like for small businesses.
We want to share some of our insights and discuss how AI is changing things for businesses, and how we’re all going to have to live in a world with AI (but not in the cool, sci-fi humans-living-with-robots kind of way). Let’s jump right in.
Quick Recap: What is AI?
Artificial intelligence is, at this point, a buzzword. It represents a technology that doesn’t quite hold up to the expectations of the dreamers and science fiction authors that popularized it, but it’s still revolutionary nonetheless. AI isn’t the robot version of Arnold Schwarzenegger from Terminator. It’s not the Great and Mighty Deep Thought from The Hitchhiker’s Guide to the Galaxy. It’s not sentient or alive or directly threatening our existence on its own, although there are plenty of dangers when it comes to AI that we’ll talk about.
The short version is this; computer scientists and researchers fed massive amounts of data into some special software. This software poured through this data. Essentially the data equated to snapshots of the public-facing side of the Internet. The software was able to read all of Wikipedia, all of the online dictionaries, tons of forums and blogs and websites, scientific articles, journals, and other publications.
The software is designed to make connections between all of this data, and with such a huge library of information, the software is able to take complex queries, interpret and understand them, and respond almost as if in its own words.
If you ask an AI “What is snow,” it comes back and gives you an almost dictionary-like definition of snow. It’s not necessarily copying the dictionary word-for-word, because it has so much information to reference that it can take bits and pieces from all of those sources to give you a partially “original” response. It’s not too far away from what a high school or college research paper might be, although usually, research papers need to cite their sources.
Why is AI so Impactful?
The way I described AI above might make it seem sort of superfluous. Granted, at that base level, it seems like you could simply use Google instead of an AI, right? After all, a Google search will typically give you the best results for a question, and in theory, could lead you to the most informative resources for your particular query.
What makes AI so powerful is that it responds more like a person, and it can interpret your intentions more like a person. It understands language and instruction better than Google, and it can take the information you give it and perform virtually any task with that information.
Here’s an example. I’ll ask ChatGPT, currently the most popular AI tool, to compose a short, original poem about snow.
ChatGPT’s response:
Beneath the hush of winter’s sigh,
Snowflakes dance from a silver sky.
Each flake a whisper, soft and light,
Weaving a quilt of frosted white.
The world transformed, serene, and still,
Blanketed plains, a quiet hill.
Footsteps fade in the crystal glow,
Lost in the dream of falling snow.
If I run that through a popular free plagiarism checker, it comes back as original. I’m no poetry expert, but this seems like it fits the criteria, and might be passable as something a human composed. Of course, the AI did this based on its understanding of language and all the information and connections it has based on the hundreds of billions of words ChatGPT trained on.
Now here’s where it gets interesting… I can ask ChatGPT something totally out of left field, with the information it just “invented” for me. I’m going to ask it to take the poem and read it to me as if the AI were a pirate.
ChatGPT’s response:
Arrr, lend me yer ears, matey, and I’ll spin ye a frosty yarn:
Beneath the hush o’ winter’s sigh,
Snowflakes dance from a silver sky.
Each flake a whisper, soft an’ light,
Weavin’ a quilt o’ frosted white.
The world transformed, serene an’ still,
Blanketed plains, a quiet hill.
Footsteps fade in the crystal glow,
Lost in the dream o’ fallin’ snow.
Arrr, can ye feel the chill run through yer timbers? A fine wintery tale fer pirates an’ landlubbers alike!
The remarkable thing about AI is that it continues to learn as you use it. It doesn’t just regurgitate information, it can continue to layer instructions over top of instructions all while interpreting what you want it to do.
Our example here doesn’t really have a good use case for most businesses, but the implications of what AI brings to the table are clearly impressive when you break it down like this.
AI Isn’t Just Text
If you’ve been using social media over the last year, you’ve probably run across images that might seem a little off or too good to be true. We could write an entire blog post about generative AI for images, as that’s a rabbit hole that goes deep. The point is, that virtually any information that humankind has a large quantity of, can be fed into an AI, and then that AI can create derivative content that is seemingly unique and original.
That goes for images, music, video, games, code, statistics, and all the way up to predicting complicated scientific models.
The Dangers of AI
Artificial intelligence, in its current form, has only been around for a few years, and it has already proven itself to have a wide range of harmful uses.
For instance, cybercriminals and scam artists can use AI to compose increasingly effective phishing emails and social media scams. There is evidence that AI has been used in such scams including the infamous grandparent scam, where scammers call parents and grandparents claiming that a child or grandchild has been incarcerated, trying to scare them into forking up bail.
Social media sites like Facebook and X have been flooded with AI-generated slop; images and text designed to get people to engage or emotionally respond to that isn’t created or posted by a human.
Deepfakes, a type of scam where video content is created that convincingly looks like a real person in order to defraud, deceive, or humiliate them, are made easier by AI. There are online services that will use artificial intelligence to create pornographic videos of someone simply by providing a few photos of them.
In other words, there is a massively toxic side to this technology, and we’re only just beginning to see how bad things can get.
AI Has Been Overpromised
Beyond the sketchy use cases, the AI industry is operating on some of the same principles of a snake-oil salesperson. This isn’t even anything new. Around 50-60 years ago when scientists and eventually businesses started to take advantage of early computing, there had been the thought that eventually, computers would be able to replace the average worker.
Even back then, the advent of the computerized employee was constantly being promised as something that is merely 5-6 years away. Predictions that workers would be replaced by intelligent machines would find their way into business and scientific journals, and while we’ve all become more effective and productive thanks to the technology we use, these systems don’t replace us. We’ve seen this 5-6 year promise come back a few times over the last few decades, and the most recent recurrence is now with AI.
While AI is a powerful technology, take all of the promises of a completely autonomous future with a grain of salt. We’ve got a long way to go in most cases before we let the robots serve us.
So why are AI companies hyping up the idea that humans would be replaced by their products?
You probably wouldn’t be too surprised if the answer was to appease their shareholders, and that’s certainly a part of it. There’s also a sort of cult-like mindset that some of these tech leaders tend to share. The linked article is sort of historical proof. The Rabbit R1 AI gadget the journalist mentioned turned out to be a huge disappointment and simply isn’t capable of everything that the developers promised.
Every software product these days is starting to boast AI as a feature simply because it looks good on a marketing slick. Many products and services are simply renaming features they’ve had for years and calling it AI. The addition of AI can be an easy excuse for some software developers to increase their pricing before anyone gets a chance to say “Thanks, but I didn’t need AI in this app.”
At the same time, we’re seeing commercial content and entertainment that is produced using AI. Both a recent Toys R Us ad as well as the new Coca-Cola ad use AI-generated imagery, and while most will just shrug it off, the ads are both a little on the unsettling uncanny side of the valley, especially when you realize these big corporations opted to pay fewer creative people to get the work done.
It’s hard to segue into saying AI isn’t replacing humans, but personally, I’d rather live in a world where AI replaces the tedium as opposed to the creative and artistic.
The Incredible Use Cases of AI for Maryland Businesses
It’s not all doom and gloom though. Small business owners and professionals should be curious and experimental when it comes to AI. With so many use cases, this technology can bring a lot of benefits to businesses. Here are some examples that we’re excited about:
AI with Modern Physical Security
Some modern surveillance systems, especially those designed for schools, universities, and larger facilities, can utilize artificial intelligence to spot potential threats faster than a human might. For instance, security cameras can detect smoke or vapors coming from a restroom in a high school and alert staff. Facial recognition combined with AI can help track specific subjects in an emergency to help authorities make faster decisions on how to handle a problem.
AI with Business Intelligence
Business Intelligence is where you take all of the data your company collects and analyze it to help make actionable decisions. Modern BI tools use AI to help analyze the data, making the process even more powerful, and also making it faster for decision-makers to get the insights they need.
AI as the Personal Assistant
Nearly everyone has some mundane tasks that they could probably automate or at least simplify, either with AI or with a specific application or integration. AI can help summarize meetings, emails, and other long-form content. It can help perform tasks in spreadsheets or help you reword a social media post. Since a lot of AI tools like ChatGPT are simply learning from Internet-based content, sometimes replacing a Google search with ChatGPT can be a shorter path to an answer you are looking for (just be sure to fact-check when necessary).
There are even integrations that tie AI solutions into the applications you use every day, such as Outlook, Slack, or Teams.
AI in Human Resources
Artificial intelligence can be used for screening resumes, ensuring that only applications with the right credentials get through. Businesses can use AI for employee onboarding and engagement as well, and many HR-related applications include AI features.
AI and the Supply Chain
Artificial intelligence can be used to manage inventory, logistics planning, and demand forecasting. This can reduce the workload and ensure that your staff is focused on the customer experience and quality control.
AI in Cybersecurity
Remember when we mentioned that cybercriminals are utilizing artificial intelligence? Modern cybersecurity platforms can use AI to stay ahead of the bad guys. This results in faster response rates and better overall visibility when it pertains to the safety of your network and your data.
AI is Here, and Dresner Group is Here to Help Your Business
If implementing artificial intelligence into your organization seems like a daunting task, you aren’t alone. We’ve been among Maryland’s top technology companies since 2002. We can help your business make sense of emerging technologies, and everything we do is security-driven with the goal of protecting your business, your staff, and your customers.
If you need help with your business technology, give us a call at (410) 531-6727.
