New news about DeepSeek and Qwen 2.5-Max AI Models. All I’m hearing is “My AI model Can Beat Up Your AI Model!”
Sorry for any burst bubbles. That’s all I hear when I read the hype about the DeepSeek and Qwen 2.5-Max AI Models. We’ve heard this song before. Alibaba released Qwen 2.5-Max, supposedly unseating DeepSeek’s seventy-two hour legacy as ‘Best AI model evar!’? Sorry folks, this is starting to sound dumb – can I recommend taking a step back?
Here’s the deal: AI models are disrupting many markets and newer models mean more efficient ways to get the same results using less power. Yes, these models are fascinating and outperforms ‘Microsoft-backed OpenAI’s o1 in AIME.’ The question simply is – so, what? What will these models help us DO?
Anyone help me with that? I mean, we’ve been using AI in various forms since the 90s. I’ve lived through – let’s see – four tech hype/bust cycles so far (PC era, Dot-Bomb, social media / mobile devices, and IoT). This song sounds so familiar, I’m starting to hum the words. Just as Dot-Coms, mobile devices, and IoT were Wall Street darlings a few years ago, LLMs and AI models are the new ‘IT Girl’ of Wall Street in 2025.
So what? Most of the speculation fails to connect the dots on what these super-duper models are supposed to be doing for us. Just as dot com companies could find themselves with $150M in funding before they had a business plan, AI models are massless centers of gravity. ‘Your car has a 350 V8? Ooh, but I have a 6.6 liter engine!’ Like, awesome – what’s that going to do *for* me?
Most of the AI model signal contains those important ‘missing stairs.’ How many more days until we find ourselves laughing at AI’s speculative bubble? The lyrical nature of these hype/bust cycles means we’re still climbing the hill toward ‘facing reality.’ The next step – after people lose trillions of dollars – will be the market opportunity and sustainable market. Which part of this process do we want to be a part of?
The ‘my AI model can beat up your AI model’ talk from DeepSeek Qwen 2.5-Max AI should scare everyone. Human beings – for all their cynicism – are gullible creatures. I think it’s time to revisit Steve Jobs’ famous 1997 speech about the importance of starting with the customer experience and working backwards. See if you can find any justification for caring about the technical benefits of one AI model versus another in this impromtu TED talk:
Yes, yes – we know: Steve Jobs was Steve Jobs. That’s not the point. What *is* the point is what we’re learning from guys like him who understood what it meant to build innovative products for people to use every day.
So what will the customer / user experience be from using Qwen 2.5-Max or DeepSeek? Even AI doesn’t know. “These advancements in AI technology have the potential to make various aspects of our lives more efficient, accessible, and informed,” sez Perplexity. “However, it’s important to note that the actual impact on quality of life will depend on how these technologies are implemented and made available to the general public.”
Exactly.
The only people getting rich off of AI model news seems to be the media / narrative complex, and stock trading platforms. Does that sound familiar? It should – ‘during the 1849 gold rush, those who made the most money sold picks and shovels to miners.‘ This new update means DeepSeek may become an investment DeepSink. Who knows how long Qwen 2.5-Max will last before being supplanted by another model?
So when it comes to DeepSeek and Qwen 2.5-Max AI models, I’m glad you can say ‘my AI model beat up your AI model.’ Good for you, enjoy whatever gold star or participation trophy or Wall Street bump you get. Nobody’s *quite* sure what these models are going to do for us, and until you’ve answered that question, I’m keeping my hype to myself.
