At the beginning of the summer, I walked into G2’s corporate headquarters in Chicago, wide-eyed and nervous about whether I had gotten the business casual outfit right.
I was thrilled that I landed a solid internship in the tech industry. But, upon arriving, I realized I had some learning to do. Since then, I’ve learned more about how the corporate world has been affected by the artificial intelligence (AI) boom, and why it matters for everyone regardless of what position or field they’re in.
Here’s what I wish I had known about AI and technology prior to starting my summer internship.
AI has been around for decades, but it has only been in recent months that it has truly begun to take off.
But what’s the fuss all about? Why should you even care about AI? First, let’s define some of the basic terms.
HCLTech defines AI as “the science of making machines that can think like humans. It can do things that are considered ‘smart’.” They continue, “AI technology can process large amounts of data in ways, unlike humans. The goal for AI is to be able to do things such as recognize patterns, make decisions, and judge like humans. To do this, we need lots of data incorporated into them.”
Kabir Sidana of Medium wrote that “the goal of AI is to mimic human intelligence in order to enhance efficiency and reduce human error.”
Machine learning (ML) is a subset of AI and involves the idea of a computer system being able to create and learn new algorithms autonomously.
Traditional computers follow an A to B format, which means they do what the creator has programmed them to do. However, ML can learn new processes and adapt to new problems on the fly.
In short, AI is the what (a computer that thinks like humans and can adapt), while ML is the how (the algorithms that detect and analyze patterns in a variety of fields).
For example, programmers don’t map out every single scenario a self-driving car might face. Instead, its system is trained to learn and make decisions on the fly.
First created in 1966 as a chatterbot (later dubbed chatbot), a chatbot is a predictive, conversational AI computer program designed to simulate human-like dialogue.
ChatGPT is maybe the most well-known and current example of an AI chatbot, but Google’s Bard and Microsoft’s AI Bing are in hot pursuit of gaining some of the market share.
Large language models (LLMs) are another form of predictive, conversational AI that is trained through data input/output sets. They are predictors, meaning whatever data is fed into the LLM is deemed by the program to be accurate. The amount of data that is being fed into these predictive programs can reach upwards of trillions of data points (also known as parameters).
For example, I used Google’s LLM model, Bard, and typed: “For breakfast today I ate…” and it responded with “a bowl of oatmeal,” “two scrambled eggs,” and “a bagel with cream cheese.” This happened because Bard previously learned that these dishes are typically eaten during breakfast time.
A major concern with LLMs is that the data being ingested into them can be unknowingly biased or inaccurate. This has allowed for some responses to be incorrect, ambiguous, and even offensive.
Currently, it appears the goal of building LLMs is not so much about making them larger with more data points, but instead, much smaller and more focused on a certain business.
This is cheaper, faster, and more accurate as the ingestible data can be authenticated prior to feeding it into the program.
Natural language processing (NLP) refers to a computer learning to understand and process spoken words the same way humans can. It takes the rules and foundation of language and combines it with a vast amount of inputted data to begin to process a natural language.
This principle is how we have voice-operated GPS systems, text-to-speech options, customer service chatbots, and more. All of these things are designed to quicken business processes, increase employee productivity, and allow customers to get accurate results faster.
Deep learning (DL) is a subset of ML that deals with larger-scale problems.
These programs are able to run multiple computations simultaneously, allowing for faster results. Many DL programs can, like ML systems, create new algorithms without the aid or guidance of humans. The programs expand their breadth of knowledge and aid us in new and innovative ways across healthcare, social media, finance, cybersecurity, and many more domains.
At its core, it’s MLg but for larger and more complicated problems. Learning, as it goes, it can store huge amounts of information to further learn and develop in a way that will be helpful for humans.
So, when did AI get its start?
The origin of AI occurred in the 1950s with Alan Turing, the father of modern computers. In 1950, Turing published a paper titled “Computing Machinery and Intelligence,” which focused on the idea that, if humans use stored information to solve new problems and make decisions, what holds back a machine from doing the same thing?
Sadly, computers back then were expensive and slow. And instead of storing commands, they were only executing them, thus prohibiting them from learning and analyzing as Turing envisioned. With time, however, computers grew in capability and memory while simultaneously shrinking in size and price.
In late 2022, OpenAI released a groundbreaking product: ChatGPT, an AI chatbot that specializes in NLP. Four days after launch, they surpassed one million users, and a month after that, experts estimated ChatGPT had amassed about 265 million unique users.
For reference, it took TikTok nine months to accumulate 100 million monthly active users, and Instagram almost two and a half years to get to that point.
Companies around the world were scrambling to stay on top of the growing demand for AI. Soon, major companies across industries were announcing the use of AI to streamline their business processes.
For example, Microsoft announced shortly after ChatGPT’s rise that they had partnered with OpenAI and agreed to invest 10 billion dollars into the research and development of AI. Other major companies followed suit, and not all of them were major technology companies like Microsoft; some weren’t even in technology at all.
Companies are always looking for ways to gain an edge over their competitors. AI is a powerful tool that can help businesses do just that.
AI can help increase business efficiency and size, allowing companies to combine human and artificial brains to maximize output and value. It can also be customized to a business’s individual needs and is a very cost-effective way to streamline a business model.
In April of 2023, EY surveyed over 250 leaders in the technology sector.
of respondents said they were in the process of exploring new ways to implement some version of AI into their organizations.
Source: EY
G2 isn’t far behind, either.
A few months ago, G2 released our very own iteration of an artificial chatbot called Monty. Quite simply, Monty allows software researchers to ask what kind of services they’re interested in. Monty then, in a matter of seconds, provides a list of suggestions.
Here’s an example of a search someone might make:
Pretty cool, right?
Here, AI is smoothing out G2’s business process. According to Tim Handorf, one of G2’s co-founders, implementing AI in G2’s business processes helps “guide users to the ideal software solutions for their unique business needs.”
All in all, AI is the future of business. It is the union between man and machine that allows for a business to scale, grow, and succeed in ways that have never been done before.
Richard Baldwin, an economist and professor at the Geneva Graduate Institute in Switzerland, says, “AI won’t take your job. It’s somebody using AI that will.”
By using AI effectively, we as a society will see a surge in productivity and output, all in all ushering forth a new generation built on the backs of hard work coupled with streamlined processes only available through AI
AI has a wide range of potential. From a personalized education system that increases in difficulty when the student is ready for the next level all the way to an AI system that finds Waldo faster than any human could, the capabilities are endless.
By applying ML to our society, we will see more positive growth results in how we utilize technology, no matter the industry. These days, it’s not enough to just use AI – you have to embrace it.
Just like us, artificial intelligence never stops learning. Learn more about how these bots are using reinforcement learning to finetune their skills.
Learning to use generative AI tools can prove daunting for the uninitiated. Between the...
by Christophilos Theodosis
After passing the bar in 2016, I transitioned from courtroom advocacy to Senior Corporate...
by Andrew Stevens
The legal industry presents a unique contrast: many of its principles have remained the same...
by Eunice Buhler
