I was recently asked by a client to talk to their team about AI. What it is, what it means, etc etc. Naturally, I turned to ChatGPT and Bing Chat to write the presentation for me.

I plugged in their loose question guidance to generate an outline and then fed each part of that into one of the chatbots to see what they spit out. Any time I used Bing Chat I asked it to research the industry or something similar first.

So, what does AI think about AI?

What Is AI?

AI is all about making machines, like computers or robots, do things that normally require human intelligence. Humans are really good at things like recognizing faces, understanding language, and making decisions based on complex information. But with AI, we can teach machines to do these things too.

How Does AI Work?

AI is all about using algorithms, which are like sets of instructions, to help machines learn and make decisions on their own. These algorithms can analyze data, identify patterns, and even make predictions based on what they've learned.

There are different types of AI, like machine learning and deep learning, but they all have one thing in common: they use data to learn and improve over time. That means that the more data a machine has to work with, the smarter it can become.

AI is already being used in all sorts of ways, from helping doctors diagnose diseases to powering voice assistants like Siri and Alexa.

What Is Generative AI?

Generative AI is a type of artificial intelligence that is capable of creating new and original content, such as images, videos, music, or text, based on patterns and rules learned from existing data. Generative AI algorithms use complex mathematical models, such as neural networks, to generate new content that mimics the patterns and characteristics of the training data.

Of course, there are also concerns about the potential misuse of generative AI, such as the creation of deepfakes or other forms of misinformation.

What's a Challenge for AI Adoption?

One of the biggest challenges is data privacy and security. As AI systems collect and analyze large amounts of sensitive information, there is a risk that this information could be misused or stolen. Another challenge is that AI systems are only as good as the data they are trained on. If the data is biased or incomplete, then the AI system will produce biased or incomplete results.

How Will AI Impact Marketing?

Targeted Ads

By analyzing data on buyer behavior and creating personalized marketing campaigns that target specific demographics.

Past Sales Future Wins

AI algorithms can identify patterns and trends that help target marketing efforts more effectively.

Predicting The Future

Predictive analytics can enable digital marketers to make better decisions about how to allocate their resources and invest in new initiatives.

Increased Efficiency + Reduced Costs

Via task automation, predictive analytics, and early issue identification.

I mean, not bad really. Various queries got really repetitive in the "how will AI impact this industry / marketing," but nothing crazy. Of course, maybe this is all a feint by the chatbots so we don't suspect their true power and motives.

This was the point in the presentation where I revealed the joke and got on my soapbox.

I am of the group that believes...

Current AI = Infinite Interns

Our current models are like having an army of super-fast interns at your disposal, and you should approach them as such.

According to AI, the difference between it and an intern is:

It can perform tasks faster and more accurately than a human intern, especially for tasks that involve large amounts of data or require repetitive actions.

It doesn't get tired or bored, so it can work 24/7 without breaks or distractions, while an intern may need breaks and time off.

It can learn and improve over time as it processes more data, while an intern may have a fixed set of skills and knowledge.

It can handle complex tasks that may be too difficult or time-consuming for an intern, such as analyzing vast amounts of data or performing complex calculations.

The AI Formula

Pattern Recognition + Pattern Generation

Like most systems, you get out what you put in. In the case of generative AIs that means more of the same. A really good rendering of the internet culture version of the thing you asked for.

The large language models (LLMs) currently taking the world by storm (GPT, etc) are really just turbocharged predictive text models. If it sounds human, it's because it was trained on words written by humans.

The near term impact heralded by the chatbots is right:

Increased Efficiency + Reduced Costs

What could you do with infinite interns?

I'll leave off with this parting thought:

The Future is Centaurs

If AI gives you more of the same, how do you stay different?

If everyone uses the same AI, how do you do something different?

If every experience is AI, how do you stay human?

It’s not AI or humans…
It’s AI and humans

Meanwhile, in robots:

• It's Google I/O, which means Big G has announced one billion things with AI. Including Search Chat* and its response to Open AI's GPT models: say hello to PaLM 2
• Chip Wars, reignited: Microsoft is getting in on the custom silicone game with an in-house design tailor made for AI & LLMs (chips are still one of my main areas to watch).
• An excellent AI thought experiment: envisioning LLMs as vast cities. "Think of it as the opposite of the Turing test setup: instead of putting a machine in a context where you can’t immediately tell whether or not it’s a human, build it in a way that’s inescapably, viscerally machine-looking."

*Don’t worry, Google Search Chat will still have ads. I’m guessing this marks the end of keywords. Not now, but soon. We’ve been on a slow march to their inevitable death, but now we’ll actually start to see what the post-keyword future looks like.

My biggest question: will Performance Max campaigns be the only ones eligible to appear in AI-powered Google Search Experiences?