Who’s Who in the AI Zoo?
Every day, the AI landscape shifts, with new tools, faster models, and more hype.
If you’re like me – determined to stay up to date with AI advancements – you may be feeling a sense of overwhelm.
Don’t despair. You’re not alone – I’m right there with you – and I’m introducing this practical AI series to share learnings, help you stay current and cut through the chatter.
I didn’t always use AI in my daily workflow. But over time, the flood of announcements and product launches led me to a simple decision: keep up or risk falling behind. Each day, I settle in at my desk and engage with AI, testing various tools to help with tasks and strategies across my workloads.
For years before the emergence of ChatGPT, the potential of AI was exciting, but lacked wide-scale adoption. Now we’re in a new era, the beginning of “AI everywhere” – where tools like Cursor and Google’s Agentspace are redefining productivity, offering exciting new frontiers for would-be users.
With the promise of AI finally becoming reality, knowing where to start – or how to level up – still isn’t easy.
As AI capabilities rapidly expand, keeping up may feel like an impossibility. There’s a good chance you’re still trapped at the starting line, along with many AI users who are stranded in the trust gap, unclear on use cases, or dabbling without any clear return on investment.
So, here’s my promise: Each month, I’ll break down what’s happening in a specific area of AI and provide reviews of major AI categories. We’ll explore LLMs, image and video generation, open-source tools, enterprise AI, agents, developer tools, and aggregators – and we’ll try to make sense of the lingo.
In the Beginning, There Were LLMs
First off, an LLM (Large Language Model), vastly simplified, is a prediction engine trained on massive amounts of text data to generate human-like responses to queries. The AI itself does not understand what it says, rather, it uses pattern recognition to mimic what a response would likely be. There are many practical uses of LLMs, and each company has a different take on how to build and present their product.
ChatGPT
Most of you have used ChatGPT, and there’s a good chance you’re quite proficient with it. The early market leader has cemented itself in our minds as AI for everything. Users have found varying levels of success managing communications, doing basic research, and creating content. ChatGPT really stood out from the crowd with its ability to create natural sounding language and adapt to varying writing styles. Its ease of use and drive to be a market leader has meant it has kept up or led the pack with innovation and providing features.
Tips for Use:
Talk to ChatGPT like you would a person, but make sure you’re specifying what you want. The more leeway you give the tool, the more liberty it takes when responding. The popular “act as if you’re a professional in a given field” is a useful shortcut if you don’t have specific enough instructions.
An AI’s ability to recall information is limited by its ‘context window’ — the amount of text it can process at once. ChatGPT doesn’t have the lead when understanding long conversations or large documents or reference materials. Keep your conversations short or refresh your instructions occasionally as ChatGPT starts to get confused after too much time.
Lastly, I strongly recommend not using ChatGPT as a replacement for search. With OpenAI admitting recently that their models are increasingly prone to hallucinations, Google’s Gemini is a better option for search.
Google’s Gemini
Although there hasn't been as much fanfare for Google's LLM, Gemini, we've seen a heavy focus on AI products from the Google team. Gemini itself boasts impressive stats, having one of the largest context windows, early integration across the Google suite of products, and Google search built-in.
Tips for Use:
With the ability to handle large context windows and stay more accurate, Gemini is much more proficient at finding correct answers to queries. Although still vulnerable to hallucinations and recalling inaccurate information, Gemini is roughly 40% better than ChatGPT at staying accurate on specific benchmarks. So, upload your documents, websites, and the context of your question in your chat before you start, as it can handle roughly 680% more ‘context’ than ChatGPT.
Claude
Anthropic’s Claude.ai has flown under the radar for many AI users simply because many don’t understand why they should give it a try. Claude stands out with its self-correction built-in and how it manages data. Built with a more business-focused offering, Claude is considered to be more secure and friendly when managing sensitive data. Although all LLMs mentioned here have coding capabilities, Claude Code stands out as a leader, so much so, that it underpins many of the coding tools available on the market.
Tips for Use:
While Claude works well for writing, accuracy, and writing code, it also seems to handle screenshots and data better than most other LLMs. This means if you have a report you’re working on, try seeing if Claude can visualize it for you. If you’re trying to understand a spreadsheet, Claude can break down the data and help understand it, visualizing how the AI comes up with an answer. Make sure to take advantage of the ‘Projects’ function when dealing with large documents or groups of documents. This feature allows users to keep those documents available to the LLM without wasting the available context window.
Special mentions
While Microsoft's Copilot, X's Grok, DeepSeek, and Meta's Llama all warrant mentions, they do not offer the same user-friendly value as the top three LLMs discussed here. Copilot excels at tasks like taking meeting notes, and both DeepSeek and Llama can be run locally for free. Grok, on the other hand, is known for its unique ability to engage in arguments with its founder. However, the specific advantages of these platforms may not be as readily accessible or beneficial to the average user.
The AI landscape is evolving rapidly, and staying current might feel impossible. But remember, the goal isn’t mastering everything overnight — it’s continued experimentation, thoughtful practice, and consistent effort. Each month, we’ll explore the landscape together, uncover new use cases, deepen understanding, and help you stay ahead of the curve. Until next time, stay curious and keep experimenting — your AI journey has only just begun.
Every day, the AI landscape shifts, with new tools, faster models, and more hype.
If you’re like me – determined to stay up to date with AI advancements – you may be feeling a sense of overwhelm.
Don’t despair. You’re not alone – I’m right there with you – and I’m introducing this practical AI series to share learnings, help you stay current and cut through the chatter.
I didn’t always use AI in my daily workflow. But over time, the flood of announcements and product launches led me to a simple decision: keep up or risk falling behind. Each day, I settle in at my desk and engage with AI, testing various tools to help with tasks and strategies across my workloads.
For years before the emergence of ChatGPT, the potential of AI was exciting, but lacked wide-scale adoption. Now we’re in a new era, the beginning of “AI everywhere” – where tools like Cursor and Google’s Agentspace are redefining productivity, offering exciting new frontiers for would-be users.
With the promise of AI finally becoming reality, knowing where to start – or how to level up – still isn’t easy.
As AI capabilities rapidly expand, keeping up may feel like an impossibility. There’s a good chance you’re still trapped at the starting line, along with many AI users who are stranded in the trust gap, unclear on use cases, or dabbling without any clear return on investment.
So, here’s my promise: Each month, I’ll break down what’s happening in a specific area of AI and provide reviews of major AI categories. We’ll explore LLMs, image and video generation, open-source tools, enterprise AI, agents, developer tools, and aggregators – and we’ll try to make sense of the lingo.
In the Beginning, There Were LLMs
First off, an LLM (Large Language Model), vastly simplified, is a prediction engine trained on massive amounts of text data to generate human-like responses to queries. The AI itself does not understand what it says, rather, it uses pattern recognition to mimic what a response would likely be. There are many practical uses of LLMs, and each company has a different take on how to build and present their product.
ChatGPT
Most of you have used ChatGPT, and there’s a good chance you’re quite proficient with it. The early market leader has cemented itself in our minds as AI for everything. Users have found varying levels of success managing communications, doing basic research, and creating content. ChatGPT really stood out from the crowd with its ability to create natural sounding language and adapt to varying writing styles. Its ease of use and drive to be a market leader has meant it has kept up or led the pack with innovation and providing features.
Tips for Use:
Talk to ChatGPT like you would a person, but make sure you’re specifying what you want. The more leeway you give the tool, the more liberty it takes when responding. The popular “act as if you’re a professional in a given field” is a useful shortcut if you don’t have specific enough instructions.
An AI’s ability to recall information is limited by its ‘context window’ — the amount of text it can process at once. ChatGPT doesn’t have the lead when understanding long conversations or large documents or reference materials. Keep your conversations short or refresh your instructions occasionally as ChatGPT starts to get confused after too much time.
Lastly, I strongly recommend not using ChatGPT as a replacement for search. With OpenAI admitting recently that their models are increasingly prone to hallucinations, Google’s Gemini is a better option for search.
Google’s Gemini
Although there hasn't been as much fanfare for Google's LLM, Gemini, we've seen a heavy focus on AI products from the Google team. Gemini itself boasts impressive stats, having one of the largest context windows, early integration across the Google suite of products, and Google search built-in.
Tips for Use:
With the ability to handle large context windows and stay more accurate, Gemini is much more proficient at finding correct answers to queries. Although still vulnerable to hallucinations and recalling inaccurate information, Gemini is roughly 40% better than ChatGPT at staying accurate on specific benchmarks. So, upload your documents, websites, and the context of your question in your chat before you start, as it can handle roughly 680% more ‘context’ than ChatGPT.
Claude
Anthropic’s Claude.ai has flown under the radar for many AI users simply because many don’t understand why they should give it a try. Claude stands out with its self-correction built-in and how it manages data. Built with a more business-focused offering, Claude is considered to be more secure and friendly when managing sensitive data. Although all LLMs mentioned here have coding capabilities, Claude Code stands out as a leader, so much so, that it underpins many of the coding tools available on the market.
Tips for Use:
While Claude works well for writing, accuracy, and writing code, it also seems to handle screenshots and data better than most other LLMs. This means if you have a report you’re working on, try seeing if Claude can visualize it for you. If you’re trying to understand a spreadsheet, Claude can break down the data and help understand it, visualizing how the AI comes up with an answer. Make sure to take advantage of the ‘Projects’ function when dealing with large documents or groups of documents. This feature allows users to keep those documents available to the LLM without wasting the available context window.
Special mentions
While Microsoft's Copilot, X's Grok, DeepSeek, and Meta's Llama all warrant mentions, they do not offer the same user-friendly value as the top three LLMs discussed here. Copilot excels at tasks like taking meeting notes, and both DeepSeek and Llama can be run locally for free. Grok, on the other hand, is known for its unique ability to engage in arguments with its founder. However, the specific advantages of these platforms may not be as readily accessible or beneficial to the average user.
The AI landscape is evolving rapidly, and staying current might feel impossible. But remember, the goal isn’t mastering everything overnight — it’s continued experimentation, thoughtful practice, and consistent effort. Each month, we’ll explore the landscape together, uncover new use cases, deepen understanding, and help you stay ahead of the curve. Until next time, stay curious and keep experimenting — your AI journey has only just begun.
.png)
