The European Union announced this week at the AI Action Summit in Paris that it will invest two hundred billion Euros in the development of AI. Curious clicking on the link leads directly to a deleted YouTube video: 'Video removed by the uploader'. These brainiacs are going to invest two hundred billion Euros of taxpayer money in AI?
One striking aspect of the story, because serious plans are as yet unobtainable, is the creation of 'AI Gigafactories', or large-scale data centers to serve as the backbone for European AI development. When politicians start spouting texts about "hundreds of billions of investments" and empty phrases like "AI Gigafactories," because data centers are apparently not sexy enough anymore, it is advisable to be vigilant.
Of course, the European rhetoric is a reaction to the ambitious American Stargate project. That too is weighed down by a Boy Scout objective like "to build and develop AI - and specifically AGI - for the benefit of all humanity."
The communique states that priorities include “ensuring AI is open, inclusive, transparent, ethical, safe, secure and trustworthy, taking into account international frameworks for all” and “making AI sustainable for people and the planet”.
It is as if miss World and Buzz Lightyear were handing in a homework assignment together.
The Guardian wrote up a clear summary of the AI summit, with three things standing out: first, the global recognition that AI is having a huge impact on society and the economy; second, that developments in AI are accelerating; and, unfortunately, third, that there is no consensus on how to regulate developments internationally.
The fear among entrepreneurs in Europe is that bureaucrats without substantive expertise will distribute the planned budget, which will result in wasted money and slow implementation.
Smarter European approach: embrace open source AI
A better approach would be to not simply spend these funds on infrastructure or vague programs, but to invest in AI companies working with open-source technologies, not based on but inspired by China's DeepSeek. By starting with a fully open-source codebase, including transparent training data, the EU can build an AI ecosystem that is widely accessible to large companies, startups, researchers, businesses and hopefully even individual developers.
The most practical approach would be the creation of a fund to invest in AI applications that build on this open-source base. This would ideally be done in partnership with existing investment funds in the market to avoid wasting taxpayer money, rather than a top-down model in which the EU itself tries to drive innovation.
The current trend within AI shows that most investment is going to large language models (LLMs), with companies like Meta and Microsoft spending tens of billions a year on AI development. This means that if Europe is not more strategic with its investment, it risks remaining behind.
Focus on open-source AI and a smart investment model rather than a purely infrastructure-driven approach could yet help Europe achieve a competitive and sustainable AI ecosystem. But if the strategy is not sharply translated into tactical and operational decisions soon, this historic opportunity will get bogged down in inefficiency and political rhetoric.
Elon Musk's OpenAI bid not for real
Elon Musk has announced his intention to make a nearly $100 billion bid for OpenAI, but the question is whether this is a serious acquisition proposal or a strategic move to thwart his archenemy Sam Altman. Musk, who co-founded OpenAI but later left acrimoniously, vehemently opposes OpenAI's transition from a nonprofit to a commercial company. A bid of this size would make it more difficult for OpenAI to move the shares held by the non profit organization to regular commercial shareholders.
A major complication is that Microsoft owns 49% of the shares in OpenAI, meaning Satya Nadella's company has a decisive vote in any acquisition. For Microsoft, a sale would raise nearly $50 billion, but the company also has a strategic stake in OpenAI because most of its AI infrastructure runs on Microsoft Azure. This makes it unlikely that Microsoft will stand and cheer when OpenAI is acquired, unless a deal is struck in which Musk's AI company XAI along with OpenAI becomes a major customer of Microsoft.
Remarkably, Sam Altman himself owns no shares in OpenAI, giving him little direct influence over an acquisition. This highlights OpenAI's unusual governance model, with control largely in the hands of the foundation that founded the company. Musk's bid therefore seems less a serious attempt to acquire OpenAI and more a tactical move to disrupt Altman's plans and make OpenAI's future uncertain. Surely investors will be scratching their heads before they will fork over the forty billion sought by Altman on a valuation of three hundred billion in this situation.

AI UI is horrible
You'd almost forget in all the fuss to take a good look at OpenAI's products. MG Siegler did not hold back about ChatGPT's sadly tuneful interface:
"Well, now we're up to eight options – six in the main drop-down and still those same two "left-overs" in the sub-menu. And technically it's nine options if you include the "Temporary chat" toggle."
At Google, the user interface (UI) is just as horrible. The makers of the most Spartan, and thus most successful, search engine ever, have managed to turn their ChatGPT competitor Gemini into an incomprehensible AI menu. It is downright woeful, because there are extraordinary capabilities hidden beneath this wretched interface. See, for example, how Google AI Studio phenomenally explains how Photoshop works.
So I asked Google Pro 1.5 Deep Research, what a name, to produce an investment strategy for the European Union based on literature research. A few minutes later, Deep Research produced this Google Doc. Far from perfect, but better than anything produced so far by the EU.
Ethereum under fire
Ethereum, for years the leader in the world of smart contracts and after Bitcoin the crypto currency with the highest market cap, is at a crossroads. Despite the rising Bitcoin price and optimism in the crypto market, especially since Trump's election victory, Ethereum remains far behind and is trading even lower than a year ago.

What are the causes?
- Lack of major updates: after "The Merge" (the switch from Proof-of-Work to Proof-of-Stake), there has been no new breakthrough.
- Increasing competition: Solana, Sui and Aptos are gaining ground with faster and cheaper transactions.
- Negative publicity: Ethereum founder Vitalik Buterin's recent tweet about communism and decentralization was taken out of context and caused unnecessary uproar.
Ethereum is still seen as a fundamentally strong blockchain, but it may lose more and more market share to newer platforms that are more responsive to users' current needs.
Huge livestream error, token price rises?
In the third episode of the NFA Podcast, which Nisheta Sachdev makes with yours truly, she surprised me with the news that NEAR Protocol's token price had risen after a team member accidentally shared the wrong screen of his computer during a livestream, unwittingly treating viewers to carnal intimacy of the eighteen-plus genre.
The crypto world is known for its unpredictable market reactions, but what happened next was exceptional even for crypto: the price of NEAR rose 5.6% to $3.50. While it cannot be proven that the livestream incident is directly responsible for the price increase, it again raises the question of how much influence, if any, "fundamentals" have on the crypto market?
If a blunder like this can drive up the price, it means the market is guided more by hype than by the true value of a project. Even the Tinder Swindler, infamous since the Netflix documentary, is launching his own token. It is leading to increasing frustration among professional developers and investors in the blockchain world.

GameStop considers buying crypto
GameStop, the company that was bailed out by retail investors in 2021 during the WallStreetBets revolt, is now considering investing in Bitcoin and other crypto-assets. By the way, the movie about GameStop is particularly worth seeing, with splendid roles by Pete Davidson and Seth Rogen, among others.
San Francisco overrun by startup teenagers
When incubator Y Combinator recently had a party, the platters went around with glasses of soda instead of alcohol: many startup founders were simply too young to legally drink alcohol. San Francisco's startup scene is flooded with very young AI entrepreneurs, many of whom left college to start their own companies.
The cost of university education in the U.S. has risen so much that despite the low success rate, entrepreneurship is a legitimate option. Outside the U.S., university education often remains a more logical route because the cost of a university education is much lower and the funding and exit opportunities for startups are not as great than in Silicon Valley.
That and much more in the third episode of the NFA Podcast, in which I also share how my experiment with investing one hundred dollars last February went down, exclusively in tech stocks.
For the hasty viewer and clicker
00:00 Introduction to NFA Podcast and Hosts Nisheta and Michiel
01:42 Surprising News in Crypto: Near Protocol Incident
03:53 Market Reactions and Near Token Performance
05:22 Ethereum's Market Sentiment and Fear Index
08:09 Ethereum's Performance Compared to Other Blockchains
09:29 Market Predictions and New Money Flowing In
11:35 GameStop's Potential Move into Crypto
12:42 Upcoming Launches: Tinder Swindler's Token
13:06 Elon Musk's Bid for OpenAI
14:44 The AI Summit and Global AI Treaties.
16:49 Youth and Startups: The College Dropout Phenomenon
20:44 Market Spotlight: Insights and Predictions
22:34 Investing Strategies and Personal Experiences.
24:44 Supermicro, Palantir and Nvidia
25:20 Dutch Trance NFA Podcast Theme
25:41 NFA Dutch Trance Theme Review
25:59 Indian NFA Podcast Theme
26:25 Indian NFA Theme Review