😸 How to know when you’re making the wrong product

gm legends. It’s Sunday funday.

In this edition: How Wispr learned to go with the Flow (and what you can learn about listening to your target customers), Elon gives away image-to-video on Grok, Perplexity’s bid to buy Google Chrome (and other) marketing stunts, and the most popular products that went live this week. 

Fill up your Yeti and camp out at your desk. We’ll show you the way.

P.S. Launching soon? We’d love to hear about it → editorial@producthunt.co 🫶

It’s hard to believe that X released Grok 4 less than six weeks ago. Remember the hentai-adjacent anime chatbot and the anarchic red panda? (Here’s a refresher.)

OpenAI’s launch last week of GPT-5 took some of the wind out of Grok’s not-quite-suitable-for-work sails. Despite its rocky rollout, ChatGPT sits at #1 on the App Store, while Grok has been struggling to stay in the top 5. (No shade—TikTok is #15). 

So Elon did what Elon does, and has been rustling up some attention.

No, we’re not talking about X temporarily suspending Grok’s account, which Musk called “just a dumb error.” We’re talking about leveraging one thing Grok has that ChatGPT doesn’t: image-to-video. About the same time GPT-5 came out, Musk announced Grok Imagine would be free to U.S. users “for the next few days.” It hasn’t been enough to move the needle past OpenAI, but it’s keeping Grok in the picture. 

Of course, the question for Grok users is: How long will they be able to keep using Imagine to make 12-second homages of Elon riding a horse into battle? It’s unlikely that the video generation tool will remain gratis forever, meaning that if you’ve just come for the videos but don’t want to stay for the anti-social behavior, you’ve got options.

We’ve been keeping track of text-to-video launches at Product Hunt. Here are a few:

Sora: OpenAI does do text-to-video and image-to-video within Sora, but its video generation model isn’t included as part of ChatGPT. There were some hopeful murmurings that V2 would be integrated into GPT-5, but alas, we’re still waiting to see it. Still, V1 won our Golden Kitty for AI for Video in 2024.

Runway: Gen-3 Alpha, RunwayML’s video-generation base model, took silver in the Golden Kitties last year. This year, Runway has already released Gen-4 as well as its Aleph video-editing tool. With the latter, you can add an image to influence how the video turns out, i.e., make the video take on certain characteristics of the image.

Veo: Our Golden Kitty bronze medalist last year, Veo won points for lasting longer than a minute. Another bonus: Google maximalists can use it as part of their Gemini subscriptions. Google intro’d V3 model this May with a powerful new feature competitors lack: sound. The frame-to-video feature, however, still has some kinks.

We could go on. Just this year: 

• Hedra released Character-3
• Luma announced a faster, cheaper model of its mobile-first model
• Higgsfield came out and began rapidly bolting on new features
• KLING AI launched V2.1
• And AI image pioneer Midjourney released its first video model 

By Tanay Kothari, founder, Wispr Flow

My childhood dream was to build a real-life JARVIS, Tony Stark’s AI assistant. And in 2021, it looked like I’d have my shot. 

It was in between the launches of GPT3 and ChatGPT. We were starting to see things happening with large language models, and I believed that in the next few years, we'd all be talking to our devices. But if everybody was talking to devices, there’d be no privacy and we’d be disturbing other people. 

I thought: If we could build a device that lets you communicate with everything around you silently, that would be a game-changer. 

So we spent half a year building an early version of that. It was a small wearable device that converted neural signals from silent speech into text or voice. Once we had it working in a limited capacity, we raised our first round of funding and were off to the races.

Over the next three years, we built an incredible team, with some of the best PhDs in neuroscience, machine learning, and signal processing. We used our funding to build a 40-person company, and we actually got this crazy idea to work. It was a pair of headphones that you put on, and when you silently speak, it would convert your brain signals into a voice that sounds exactly like you. 

It literally felt like magic.

But that was the problem.

AI search engine Perplexity tried to buy Google Chrome this week. The $34.5 billion bid, which some have labelled a stunt, is worth roughly twice Perplexity’s value. Perplexing? Maybe! But it also got media hits. (Probably just a coincidence that Perplexity launched its Comet web browser Thursday.)

In light of this, Aleksandar asked: “What’s the craziest thing a startup can do to get attention?”

Nika chimed in with a recent example from crypto involving, well, you’re just going to have to read it to believe it.