🏆 The AI Olympics—Who's Winning

The Street

The AI Olympics: Chatbots Ranked In Writing, Coding, Images & More

Let’s be honest: keeping up with the latest GenAI chatbots is a full-time job. Fortunately, many experts do make this their job—and we read their homework.

Inspired by real-world GenAI testing (including Claude, ChatGPT, Grok, and Gemini), we broke it down by use case—plus added new categories, provided both free 🆓 and paid 💸 options, and didn’t include any tools that stall after just a few prompts.

Here are the categories and winners:

🧠 Idea Generation & Chat

• 💸 ChatGPT-4o: Seamlessly combines text, image, and audio into one fast, multilingual brain—and responds in milliseconds like a real conversation.

• 🆓 Claude Sonnet: Calmer and more human sounding than ChatGPT—great at deep convo, curious prompts, but can get a bit wordy and change the convo unprompted.

🎨 Image Generation

• 💸 ChatGPT-4o: Beat popular AI image generator Midjourney in a photo-to-art test, less prompt engineering required to obtain good results, excellent at creating in another artist’s style.

• 🆓 Gemini: Delivers similar visuals and lets you edit parts of an image—like removing a photobomber, which ChatGPT can’t do much to its detriment.

💻 AI Coding

• 💸 GitHub Copilot Pro: Best for real-time code completion—it generates suggestions as you type, directly in your IDE, essentially embedding an LLM into your development workflow.

• 🆓 Claude Sonnet: Explains code clearly, refactors well, and writes clean functions—but you still need to paste into an IDE to run it.

✍️ Fiction Writing

• 💸 ChatGPT-4.5: Best storytelling of any model—subtle, structured, expected to hallucinate less, but lacks chain of thought reasoning.

• 🆓 Gemini 2.5 Pro: Sharp metaphors, good pacing, and great emotional depth layering of the characters.

For the Data Nerds: Real-Time AI Rankings

There are plenty of platforms benchmarking models in real time—Artificial Analysis is a favorite, with leaderboards that weigh intelligence, speed, and cost. The image below shows OpenAI, Gemini, and Grok leading the intelligence charge.

Surprisingly, Claude Sonnet scores just a 57, proving that raw intelligence doesn’t always equal usability—kind of like people, really. 😉

In Conclusion—The Best AI Is the One You’ll Actually Use

Certain LLMs shine for certain use cases, that much is clear. And in general, picking a chatbot is like choosing a messaging app—they all send texts, but the feel, tone, and features can make one your go-to. It comes down to what fits your workflow and vibe.

And with near constant releases and upgrades, keeping score is getting harder by the week.

But if you had to choose just one AI to rely on daily—for writing, ideas, research, or real conversation—experts tend to agree, there’s still nothing quite like ChatGPT.

💡How To AI: Your New Assistant

Wish you had an extra set of hands for life’s boring-but-important tasks? AI might not do the dishes (yet), but it can handle the admin so you don’t have to. Here are some clever ways to use it.

📅 Schedule Smarter: Let AI summarize meetings and takeaways and prep your day by analyzing your calendar (just don’t ask it to fake a dentist appointment to skip the team sync).

🧠 Daily Health Plan: Have an LLM plan your workouts, build a meal plan, then use OpenAI’s Operator to order your groceries—all without leaving your couch.

🎁 Thoughtful Gifter: Tell the LLM who you’re buying for (“my cousin who just had a baby and loves hiking”) and get a tailored gift list with links. No need to read NYT’s gift list anymore!

HubSpot has created a smart roadmap to help you pinpoint exactly where AI can lighten your workload.

Ready to save precious time and let AI do the heavy lifting?

Save time and simplify your unique workflow with HubSpot’s highly anticipated AI Playbook—your guide to smarter processes and effortless productivity.

Download the free guide today.

Heard in the Server Room

Copyright Office Chief After Releases AI Report, Gets Fired

Ever dropped content so hot it got you fired? Shira Perlmutter, head of the U.S. Copyright Office, now knows the feeling—she was dismissed shortly after releasing a report raising concerns about how AI models use copyrighted material. The report questioned whether more data always leads to better AI and subtly pushed back on Elon Musk’s calls to scrap IP laws (an IP enforcement office defending IP? Who would've guessed). While the fallout is still unfolding, tighter copyright enforcement could limit how companies train their models—possibly leading to less capable AI, but a win for artists seeking fair compensation.

New Pope Leo XIV Lists AI As A Top Challenge To Humanity

Pope Leo XIV isn’t wasting time, already spotlighting artificial intelligence as one of the biggest challenges facing humanity (and the Church of course). In his first big address, the new pontiff compared today’s AI boom to the Industrial Revolution, echoing Pope Leo XIII’s push for worker protections back in the 1890s. Leo XIV made it clear the Church needs to be front and center in tackling AI’s impact on dignity, justice, and jobs, signaling continuity with Pope Francis, who also called for global AI regulation.

From Crude to Code: Saudi Launches AI Power Play

Saudi Arabia is throwing its keffiyeh into the AI ring with the launch of Humain, a new state-backed company chaired by Crown Prince Mohammed bin Salman. The kingdom wants to swap oil for algorithms as part of its Vision 2030 plan, aiming to become a global AI hub with shiny new data centers and cloud infrastructure. Meanwhile, the U.S. is tuning in—Trump's Gulf tour kicks off in Saudi Arabia where AI will headline investment talks. Salesforce is already on board, pledging $500 million in AI investment in the Gulf state.

This AI Doesn’t Just See Your Face—It Sees Your Future

Researchers at Mass General Brigham just released FaceAge—an AI tool that can guess your biological age (for most of us, that’s higher than our actual age) and predict health outcomes—all from a simple selfie.

Trained on nearly 59,000 images of healthy individuals, FaceAge isn’t just scanning for gray hairs. It looks at subtle facial features that correlate with how quickly the body is aging. In clinical studies, it outperformed 10 doctors at predicting survival for palliative care patients.

Doctors are intrigued but cautious. ER physician and AI futurist Dr. Harvey Castro calls FaceAge “a quantified eyeball test,” but warns about bias in training data, ethical red flags like consent, and the awkward moment when AI informs you that you look old.

Why it matters: FaceAge could give doctors a faster, non-invasive way to assess patient risk, personalize care plans, and catch red flags earlier—especially when time is critical. It won’t replace human judgment, but it could become a powerful second opinion.