"Never forget that yesterday's assumptions are today's vulnerabilities!" - Futurist Jim Carroll
It was a wild day for AI yesterday. And it was also a wild day for stocks. The two are interrelated. You might have noticed that tech stocks took a beating.
All of this came about as the result of the release of a new model called DeepSeek.
There are a lot of human-friendly explanations as well as deeply technical treatises on what this is all about. Suffice it to say, that the nature of the underlying technology and the cost to bring it to the world changed every single assumption as to what it takes to bring a sophisticated AI model to the world. This is one of the best summary posts in what I saw being reported.
The fact is, it was a very disruptive announcement, and is a great reminder of the truth that any assumption that a business or industry might make yesterday about its future could be entirely disrupted today and tomorrow. Later in this post, I'll share a note from tech visionary Peter Diamandis which does a great job of putting in perspective this reality.
Naturally, I was curious as to whether all the hype lived up to the reality. One of the key things about the new model is not only does it cost a fraction of the cost of ChatGPT and other similar models, but it can pretty well offer much of the same performance if run locally, i.e. on your computer. Without being technical, this doesn't mean *all* computers, and the sophistication of what you can do on your machine is limited by the power of your machine. That said, I went and installed one of the smaller data sets for the model - 9GB, vs 460GB for the full model and asked it to ponder on today's quote.
What I got back was pretty sophisticated for the small size of the model on my Mac. In the screenshot, you can see the analysis it undertook while 'pondering' this question.
Does the statement 'Never forget that yesterday's assumptions are today's vulnerabilities' bear any validity? What are the underlying assumptions that business organizations might often make that can be found in this type of statement?
Once it was finished 'thinking' about the quote, it offered up this response.
The statement "Never forget that yesterday's assumptions are today's vulnerabilities" holds significant validity in the dynamic business landscape. It underscores the importance of recognizing that past assumptions can become weaknesses if they are not revisited and updated in light of changing conditions. Here's a structured breakdown:
Validity:
- Adaptation to Change: Business environments are constantly evolving, with shifts in markets, customer preferences, and technology. Relying on outdated assumptions can lead to missed opportunities or increased risks.
- Strategic Relevance: Strategies and models that were effective in the past may become obsolete if not reassessed for current realities.
Underlying Assumptions as Vulnerabilities:
- Market Stability: Assuming market conditions remain static, ignoring potential shifts that could impact business operations.
- Customer Behavior: Believing customer preferences are unchanged, failing to adapt to evolving needs or trends.
- Technological Stagnation: Overlooking advancements in technology that could disrupt current business models or processes.
- Strategic Effectiveness: Continuing with past strategies without evaluating their ongoing suitability and effectiveness.
- Process Efficiency: Assuming existing processes remain optimal, neglecting improvements despite new tools or methods.
- Skill Obsolescence: Ignoring the potential for employee skills to become outdated in a rapidly changing field.
- Risk Management: Focusing solely on known risks while ignoring emerging threats due to environmental changes.
In conclusion, businesses must regularly challenge and update their assumptions to navigate an ever-changing landscape effectively, thereby mitigating vulnerabilities that could arise from clinging to outdated beliefs.
This is a pretty good AI-generated answer to my question, from a file that takes up only 9GB on my Mac. (That's a really small file!)
Cast your mind forward - this implies that we'll soon see specific AI models that fit on our watch, in our cars, in our home appliances, and just about anywhere and everywhere. The dramatic shrinking of AI and the collapse of cost is real - and while we might not want a lot of this stuff, it's going to begin to become a part of our lives regardless.
Always remember - assumptions about the longevity of a business model, product, technology, or industry can always disappear in a moment. Assumptions are often the dangerous drug that leads us down the path of invincibility- and as history has shown, no one and nothing is ever invincible!
NVIDIA’s moat just turned into a puddle (here’s why),
Peter Diamandis, January 28, 2025
Netflix ate Blockbuster for lunch. Amazon decimated all of retail. And Uber and Zoom reinvented transport.
The disruption and reinvention of every industry is rapidly becoming the new normal.
This week, a small open-source player out of China called DeepSeek is disrupting AI giants OpenAI and NVIDIA.
But this is just the beginning! The coming industry disruptions will accelerate fast and furiously on the heels of AI, AGI, and ultimately digital super intelligence.
First the tech players, next healthcare, then financial services, and education will be reinvented. Get ready of the roaring 20’s. They are here.
Why is everyone up in arms about DeepSeek?
Here’s the data:
OpenAI was founded 10 years ago, has around 4,500 employees, and has raised $6.6 billion in capital.
DeepSeek was founded less than 2 years ago, has 200 employees, and was developed for roughly $5 million.
Here's the Disruption: While tech giants like OpenAI and Anthropic have been spending $100M+ just to train their AI models, this small 200-person team out of China built an AI system matching GPT-4's performance for 20x less money.
How did they achieve the impossible? Three Moonshot innovations that are absolutely mind-blowing:
#1. Precision Reimagined: Instead of using computational overkill (32 decimal places), they proved 8 is enough. Result? 75% less memory needed. Sometimes the most powerful innovations come from questioning basic assumptions.
#2. The Speed Revolution: Traditional AI reads like a first-grader: "The... cat... sat..." But DeepSeek's multi-token system processes whole phrases at once: 2x faster, 90% as accurate. When you're processing billions of words, this is transformative.
#3. The Expert System: Instead of one massive AI trying to know everything (imagine one person being a doctor, lawyer, AND an engineer), they built a system of specialists. Traditional models? 1.8 trillion parameters active ALL THE TIME. DeepSeek? 671 billion in total, but only 37 billion active at once. It's pure genius.
The results are staggering:
- Training costs slashed from $100M to $5M
- GPU requirements slashed from 100,000 GPUs to 2,000 GPUs
- 95% reduction in API costs
- Runs on gaming GPUs instead of specialized hardware
- They did this with a team <200 people, not thousands
But here's what makes this truly revolutionary: It's all open source.
Anyone can verify, build upon, and implement these innovations. This isn't just technological progress—it’s the democratization of AI at an unprecedented scale.
I just downloaded it on my iPhone to play with.
Think about what this means:
- The "only-big-tech-can-play" era is OVER
- Innovation barriers have been shattered
- A few good GPUs might be all you need
- The playing field has been leveled
For incumbents like NVIDIA, this is terrifying. Their business model of selling super-expensive GPUs with 90% margins? That moat just turned into a puddle.
And DeepSeek did all this with fewer than 200 people. Meanwhile, Meta has teams whose compensation alone exceeds DeepSeek's entire training budget... and their models aren't even as good.
This feels like one of those moments we'll look back on as an inflection point, like when PCs disrupted mainframes or cloud computing changed everything. The efficiency genie is out of the bottle, and there's no going back.
The question isn't whether this will transform AI development—it’s what YOU will build with this democratized technology. We're living in an era where breakthrough innovation isn't just possible, it's accessible.
Futurist Jim Carroll often likes to test and try the assumptions in any new technology before prognosticating about it.