Data has always been central to business decision-making, and most analytical workloads...

"Bloomberg Crypto" covers the people, transactions, and technology shaping the world of decentralized finance. Today's guests: Strategy Founder & Executive Chairman Michael Saylor, Vanderbilt Law School Associate Dean Yesha Yadav, and The FinTech Foundation Founder & CEO, Georgetown Law Agnes Williams Sesquicentennial Professor Chris Brummer. (Source: Bloomberg)

We launched Lakeflow Designer to general availability in June, and since then, over...

Michael Saylor, founder & executive chairman of Strategy, joins Carol Massar and Tim Stenovec on "Bloomberg Crypto." Michael Saylor says he's 'disappointed' in MSCI, which may remove his company from its indices this Friday. (Source: Bloomberg)

Alcibiades being taught by Socrates, Francois Andre Vincent (1777). Wikimedia Commons, CC BY Most students now use AI in their tertiary studies. AI can also, under certain conditions, answer at least 65% of questions correctly in STEM subjects. In the humanities, where the whole aim is critical and creative reflection, educators face a thorny question. How do you develop and measure thinking of the kind that a machine could not genuinely imitate? AI has proved it can answer questions well. But can it ask questions? And can it ask the right questions? This is the philosophical method at the heart of the work of the Ancient Greek philosopher, Plato. His Socratic method of teaching is, perhaps, the mode of thinking we need to emulate in our teaching and learning today. The great questions Plato wrote dialogues: texts that read (in nearly every case) like a conversation between Socrates (Plato’s teacher) and a partner drawn into conversation by Socrates after claiming to know something about living a good and virtuous life. Goodreads There is a dialogue for all the great philosophical questions: what justice is; what courage or love really requires; what makes a good friend; and even how we acquire knowledge. In each case the conversation is set by Socrates, who questions someone who seems to possess knowledge about a certain thing, or who has come to Socrates for advice. It is not just the content of these dialogues that has prompted centuries of discussion, but their form. Plato’s dialogues are extraordinary literary achievements, drawing on a range of genres and placing characters, historical and fictional, inside deliberately constructed settings. These are challenging and strange texts in many cases. They ask of their readers not just to read them as dialogues, but to call into question what we think we know. The Euthyphro Take the Euthyphro, the dialogue students most often meet first, which opens the sequence of dialogues dramatising Socrates’ trial and death. Socrates is on the way to his own trial, over (politically motivated) charges of failing to honour the gods of this city, introducing new divinities and corrupting the youth. He meets a man called Euthyphro, who is on his way to court to accuse his own father of murder. Euthyphro insists this is the pious thing to do. To Socrates, this is evidence that Euthyphro must know something about what piety is. Dialogue ensues. Socrates leads Euthyphro through a frustratingly methodical discussion using elenchus, also called the Socratic method. The method proceeds roughly like this. One person, Socrates, mostly asks the questions. The partner must respond to every question. A definition or principle relevant to the discussion is sought from the partner. Socrates seeks clarification, gaining assent to various propositions. Those propositions are used to show that the definition under examination is unsatisfactory. This continues until the partner flees, erupts in frustration, or admits to being in a state of perplexity, called aporia. It is this final step, the experience of aporia, that sits at the centre of the early dialogues. Aporia teaches us that we don’t know what we thought we knew, and we are all the better for finding out. After all, it is better to have knowledge of our ignorance than to be ignorant of our ignorance. This is the product and achievement of the method, though Socrates’ partners rarely saw it that way. His incessant questioning of often prominent Athenians earned him a bad reputation. Ultimately, he was sentenced to death. The Death of Socrates, Jacques Louis David, 1787. Wikimedia Commons, CC BY-NC-SA An invitation But there is more to the Euthyphro, and to all of Plato’s dialogues, than what the characters say. As modern readers of them, we are set a challenge. We can read along while Euthyphro falters at every turn, unable to show that he knows anything about piety, until he storms off and the dialogue ends. We can then close the book and go on with our lives. But to do this would be to be no better than Euthyphro who, after having his ideas of piety brought under examination, realises he must not know as much as he thought. Plato’s invitation, through the dialogical form, is to bring us into the conversation with Socrates. What do we think piety might be? Each dialogue puts us in touch with a question about an idea in this way. The dialogues ask what love, friendship, knowledge, and so on, could be. And to be properly engaged with these texts, we have to follow Socrates’ successive attempts to define the thing under examination. The aporia at the close of the dialogue is often shared, too, by readers joining this conversation more than 2,000 years later. This is one of the great lessons of studying Plato. Plato’s texts are not decrees or collections of aphorisms as they are sometimes treated to be. They are the beginning of a philosophical conversation with ourselves. They portray far more than the ideas of one Ancient Greek philosopher. They set out a way of doing philosophy, of thinking critically and creatively about who we are, our relationship to the world and to each other, and what we really know. This is not just the Socratic method, but Plato’s central teaching. Philosophical dialogue as a way of knowing. Uses today What Plato’s dialogues show is that learning takes place through the to-and-fro of reflective thought and that much of learning is really unlearning. Socrates picks his partners because they claim to know something and the discussion shows that claim to be hollow. But, crucially, the result of the discussion is to arrive at a question that is now far more refined. The question is where learning begins. To want to know something, you must already have some grip on what it is that you don’t know. You have to recognise the shape of the gap in your knowledge. If I want to understand how plants photosynthesise, for example, I must first have noticed there is a process behind the plant’s survival, something about its need for sunlight, that I don’t yet understand. The desire to know begins in acknowledging what I don’t know and this is why the aporia Plato directs at his reader is not a dead end. It is where philosophy, and all of our seeking and learning, begins. A student who asks AI to explain the causes of the French Revolution (simply because this was a question assigned to them) has asked a question, but not one that costs them anything. Nothing in their own understanding was at risk in the asking. The question that matters, philosophically, is the one that exposes a gap in the knowledge of the one who is asking – and this is precisely what AI cannot do on the asker’s behalf. AI can fill gaps. It cannot locate them for you in any way that counts as learning. This is exactly what is foreclosed when a student turns to AI before they have sat with their own aporia or confusion. AI answers before the question has been sharpened, and often before the student has worked out what they were really asking in the first place. The danger is not that students get answers. Students have always sought answers, from textbooks, from teachers, from each other. The danger is that the experience of aporia is skipped entirely and the authentic and generative question that might have led somewhere is never found. The difference between asking a question, and asking the right question, is in the proper determination of the limits of our knowledge. So, although the adage goes, “there is no such thing as a silly question,” some questions are seemingly more meaning bearing than others. They are the questions that genuinely express our wanting to know the answer. I am suggesting this is what reflective, question-posing teaching and learning could recover. To develop and assess a student’s learning today, we could find ways for students to identify and then evaluate the gaps in their own knowledge. This means, as some universities have started to do, encouraging and assessing dialogue in the classroom and utilising reflective, as opposed to purely descriptive, assessments. From here, the teacher’s task is to encourage the student to pose new meaningful questions, which are interconnected with the student’s own learning journey. Now, of course, this cannot replace every other mode of teaching and learning. The descriptive and practical training that a chemist or a mechanic receives cannot simply wait for students to ask the right questions, and explicit teaching and demonstration remain indispensable for the development of technical skill. But it would be a mistake to think only the humanities benefit from the Socratic method. Every profession contains moments where the expert must make a moral judgement and the answer cannot simply be looked up. But the deeper contribution the Socratic model makes beyond philosophy is not confined to ethics. It is the cultivation of judgement itself, the capacity to find the right question, which sits at the heart of expertise wherever it is found. Socrates made this point when he claimed his only advantage over his fellow Athenians was knowing that he did not know. We now have a tool that seems to know almost everything. The task for tertiary education is not to compete with AI at answering. It is to protect the conditions under which students still recognise what they do not know. That recognition cannot be outsourced. It is where genuine learning begins. Oscar Davis does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.

Nvidia chief executive Jensen Huang, centre, speaking at the White House. OpenAI president Greg Brockman is on the right. AP Imagine a company that sells you equipment, invests money in your business, and then rents back what you build with it. It’s your supplier, your investor and your customer all at once. That’s known as circular financing. Many artificial intelligence (AI) companies, including Microsoft, Amazon, Google and Nvidia, have adopted a version of this opaque method of financing. A chip supplier invests in a customer such as a data centre builder, the customer spends the money on the supplier’s chips, and both companies report growth. One central bank analysis found that 55.2% of investments into AI firms came from other AI firms. What’s not clear to investors or the market is how much of this AI boom is being driven by real, end-user demand — and how much might be a bubble. It all comes back to Nvidia Take Nvidia. The world’s leading chipmaker is investing in ChatGPT owner OpenAI to support data centres filled with Nvidia chips. Nvidia will invest US$30 billion in a private funding round in OpenAI and provide a guarantee of up to US$105 billion for an OpenAI data centre in Ohio, as well as investing in the company building it. That helps OpenAI to remain Nvidia’s best customer, but it also raises the question: is the chipmaker propping up the market? Nvidia has invested in other AI firms as well, including up to US$10 billion in Anthropic, the developer of Claude. It bought US$2 billion of shares in cloud provider CoreWeave to help it speed up the build-out of AI factories that use Nvidia technology. In total, Nvidia made more than 50 different investment deals in AI companies in 2024 alone. Some of those deals are staged over years or still open-ended. But these circular deals help the AI build-out move faster than it could without Nvidia’s financing. The Australian story – Firmus Data centre builder Firmus – which last week dramatically abandoned plans to list on the Australian Securities Exchange – provides the local example. Nvidia wears all three hats for Firmus: Firmus chief executive Oliver Curtis. His $$44 billion IPO failed to win investor support. Supplied Investor: Nvidia first invested in Firmus Technologies through the company’s A$330 million capital raising in September 2025 to finance data centres in Tasmania. It then participated as an existing investor in two further funding rounds, one in April 2026 and Firmus’s US$2 billion funding round in August 2026. Nvidia did not disclose how much it invested. Supplier: Firmus plans to build data centres using Nvidia’s DSX AI factory platform, which combines chips, networking and software to provide computing capacity Customer: Firmus signed Nvidia as a customer for its data centre in Batam, Indonesia, worth up to US$30 billion over six years. Meanwhile, Open AI (with its existing relationship with Nvidia) has been named as an anchor customer at Firmus’s Malaysian data centres. These circular deals also concentrate risk The same loop concentrates risk. With Nvidia acting as investor, supplier and financier across the AI sector, trouble at one firm can spread quickly to the others. It also complicates valuations. For instance, OpenAI’s method of reporting revenues has been queried. Its shares were hit last week when the Financial Times reported its annualised revenues were US$20 billion less than the company had suggested. The Reserve Bank of Australia warned in its recent Financial Stability Review that there are “pockets” of higher-risk companies, including those using opaque or circular funding arrangements, where chipmakers provide financing to the companies that purchase their products. Warning about the possibility of the AI bubble bursting, the RBA said: One possible trigger could be a shift in sentiment towards the AI investment boom, which is increasingly fuelled by expectations of sustained rapid earnings growth and a debt-financing cycle that is becoming more opaque and circular. The concern is that nearly all the gains in the US stockmarket have been driven by AI companies. On Wall Street, chipmakers have contributed around 70% of the total gains in 2026 for the S&P 500, and Nvidia is now the index’s largest single company at about 8.06% of the index. This makes the market fragile. If investors begin doubting that OpenAI and similar customers can earn enough in revenue from customers to meet their commitments, Nvidia would be hit on two fronts. First, via lower expected sales and second via its financing exposure to those same companies. That would drag down the whole market, due to its heavy weighting in the key index. Australians would feel it too. Nearly 90% of households are exposed to AI stocks, due to the US shares held through their superannuation funds. A question of timing None of this means the AI boom is fake. The issue is timing. Nvidia, OpenAI and Firmus are betting that future revenue from users and businesses will grow fast enough to pay for today’s massive data centre build-out. Until then, investors, lenders and each other’s companies fill the financing gap. There is a historical comparison with the 1990s telecom boom. In what was then called the “New Economy”, equipment makers such as Cisco Systems financed customers to buy their equipment. Too much fibre was laid too soon, and investors lost heavily when the stock market bubble burst. But that fibre lasted decades. The newest AI chips have a shelf life of only a few years. If the expected revenue and demand from business and consumers arrives late or below forecasts, shareholders and super funds will bear the losses. The authors do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.

Discussion | Link

Michael Saylor, founder & executive chairman of Strategy, joins Carol Massar and Tim Stenovec on "Bloomberg Crypto." They discuss Strategy's plan to buy and sell Bitcoin. (Source: Bloomberg)

Amber Roberts, Staff Technical Marketing Manager at Databricks, has spent the last...

Discussion | Link

Michael Saylor, founder & executive chairman of Strategy, joins Carol Massar and Tim Stenovec on "Bloomberg Crypto." They discuss Saylor's continued bullishness on the price of Bitcoin, how Bitcoin doesn't need the Clarity Act to succeed, and his disappointment in MSCI indices. (Source: Bloomberg)

Yesha Yadav, professor of Law and Associate Dean at Vanderbilt Law School, joins Carol Massar and Tim Stenovec on "Bloomberg Crypto." They discuss how sports is creating an 'attention economy' around prediction markets and the legal uncertainty still surrounding the industry. (Source: Bloomberg)

Agents got good. The hard part now is everything around them. The post Your Agent Doesn't Need More Freedom appeared first on Towards Data Science.

Disney may acquire tech from one of the first chatbot makers linked to child suicide.

Saudi Aramco’s Texas oil refinery owns some of the crude carried by a tanker that was the subject of a cyberattack en route to the US.

Hannah (Linan) Zhou, Founding Member and Head of Product at Meshy is a product leader focused on building and scaling generative AI tools for 3D content creation. At Meshy, she helps shape the company’s product strategy across consumer and enterprise applications, including 3D generation, AI agents, developer APIs, and workflows designed to integrate generative 3D into professional creative pipelines. Zhou has described her work as centered on building and scaling Meshy’s 3D generative AI…

Federal Reserve Bank of New York President John Williams said he still sees one more interest-rate hike as appropriate this year, but at the moment doesn’t believe it will be necessary for the central bank to increase rates again in 2027.

The higher prices are said to be unrelated to the Skydance merger.

An employee's AI assistant can now do more than answer a question. It can connect to company systems, install extensions, run commands and build applications. For security teams, the challenge is deciding which of those actions should be allowed—and retaining the ability to intervene when the context changes. Pluto Security is emerging from stealth with $42 million in total funding to address that problem. The company announced an approximately $34 million latest financing round led by Jump…

We are incredibly excited to announce that Databricks has made its inaugural appearance...