AI in 2025: Trying to Keep Up Without Losing My Mind

Feb 35 min read

If January and the First Few Days of February Are Any Indication, Buckle Up.

Lori spinning from all of the AI News - A GIF Created with ChatGPT

AI is Moving Fast—And It’s Hard to Keep Track

Holy Camole, didn't expect breaking AI news on a Sunday, but it's just non-stop! OpenAI just released Deep Research, an initiative focused on tackling fundamental AI challenges—because, clearly, there wasn’t enough happening already.

OpenAI introduces deep research Feb 2, 2025

Why did OpenAI Build it?

Deep research is built for people who do intensive knowledge work in areas like finance, science, policy, and engineering and need thorough, precise, and reliable research. It can be equally useful for discerning shoppers looking for hyper-personalized recommendations on purchases that typically require careful research, like cars, appliances, and furniture.

Deep Research is currently only available on the $200 a month pro plan but it's coming soon for other paid ChatGPT plans. (Sidebar: The above GIFs were created in ChatGPT simply because I was curious)

Crazy how fast things become old news

Just when you think you've caught your breath, the next advancement hits. I had a blog ready to post today recapping the insane release pace in January only to wake up to a fresh weekend release before the first business day of February. I can't even begin to imagine what the next 11 months will bring.

Tracking the Chaos: My AI Advancements Timeline

To help wrap my head around it all, I built a timeline to help visualize the pace of new developments since OpenAI released ChatGPT to the public in November of 2022. I used Perplexity to pull the key dates, Claude to help visualize the milestones and write code for an interactive map, and finally Lovable to publish it. Sounds simple, right? The fact checking of the dates was actually a nightmare. In fact, Perplexity initially used story publishing dates instead of actual release dates. Again, this reinforces why it's so important to have a human in the loop. And please know, this timeline is still a work in progress. But hopefully it gives you the gist of how fast and furious the changes are being flung at us.

Check it out here: AI Advancements Timeline.

The Competition is Intense And That’s a Good Thing

Looking at the timeline, one thing is clear and that is there is no slowing down. The competition is fierce, and I like to think that's a good thing. Every new model release, every improvement in reasoning, and every open-source alternative that enters the game is pushing the entire AI ecosystem forward. We’re in a constant cycle of innovation, and the advancements we’re seeing in just a single month used to take years.

My Go-To AI Tools—And Why I Pay for All Three

For all the new tools popping up, my go-to AI assistants are still the same core three:

ChatGPT Plus ($20/month) – My everyday workhorse. Great for writing, brainstorming, and experimenting with different prompts.
Claude Pro ($20/month) – My go-to for thoughtful reasoning, long-form analysis, and project planning.
Perplexity Pro ($200/Annual) – My research powerhouse. Real-time data, solid citations, and great for getting a second opinion.

Yes, I’m paying for all three, and yes, I think it’s worth it. Each one has strengths that complement the others, and in a world where AI is evolving this quickly, having access to multiple perspectives is invaluable. I can't tell how many times I take a conversation from one tool and then get feedback from another.

AI is Finally Showing Its Work (And That’s a Big Deal)

One of the biggest concerns with AI initially was that the information pretty much came from a black box. You have no idea how it came up with its output. Well, these new models show their work. I love seeing all the steps the models go through to generate their answers.

You can experience that for yourself with OpenAI’s new o3 models or DeepSeek.

Getting Familiar with DeepSeek

What is DeepSeek, you ask? It's the Chinese answer to OpenAI, developed at an astonishing pace that has raised questions about how quickly it was built. Some reports suggest that DeepSeek’s rapid development was fueled by learning from past AI challenges, allowing them to move faster than expected. There are also concerns about data privacy and security, with some experts warning about potential risks tied to its usage.

I downloaded the DeepSeek app but didn’t create an account.

DeepSeek in Perplexity: AI Models Within AI Models

Instead, I’ve been experiencing DeepSeek within Perplexity Pro, which stores its data in the U.S. This highlights a new trend: before, it was about choosing a tool, but now, we have so many choices within the tools themselves. It’s no longer just about picking one AI assistant—it’s about navigating the layers of options within each platform.

Perplexity Pro Model Options including Deepseek hosted in the U.S.

I’m Not an AI Expert—Just Someone Figuring It Out Like Everyone Else

I want to be clear—I’m not an AI expert. I’m someone who’s on the front lines, talking to people, seeing how they’re using AI, and figuring this out alongside everyone else. My perspective comes from daily conversations and hands-on experience, not from research labs or corporate AI strategy meetings. And that’s why I find this space so fascinating—because it’s real, it’s evolving fast, and it’s reshaping how we work and think in real time.

What’s Next? More Choices, More Competition, and Now Asking AI to Help Navigate AI

So, if this is just January, what’s next? More competition, more capabilities, and probably even more models to sort through. If you’re overwhelmed, you’re not alone. We’re now asking AI to help us choose between AI options—tools like the AI Model Matchmaker (a CustomGPT designed specifically for ChatGPT models) are popping up to help navigate the chaos. (Thanks to Liza Adams for sharing this on LinkedIn.)

AI Can Do a Lot, But Counting Is Still Not a Strong Suit

So, in an attempt to visualize this wild start to 2025, I had AI generate an image of a timeline—January front and center, followed by the next 11 months. Sounds simple, right? Well... let's just say AI-generated text and numbers are still a work in progress. Take a look at the image, and you might notice some 'creative liberties' taken with basic sequencing.

It's a fun reminder that while AI is making leaps and bounds in reasoning and creativity, some things—like counting correctly—are still a challenge. Maybe by December, it’ll have that figured out.

How Are You Keeping Up?

What are you using the most right now? Are you sticking to one AI tool or jumping between a few like I am? Let me know—I’d love to hear how others are keeping up!