How Automation Will Change Trading

🔥 Will AI Replace Day Traders? 🔥 Unbiased Evidence. Real Data. No Hype. 🔹 Algorithms already account for 60–75% of all US equity trading 🔹 Study of 81,300 traders across 28.5 million trades: AI only outperformed humans 54.5% of the time (small edge) 🔹 Hybrid trading ( Human + AI) reduced extreme human errors by 90% and AI errors by 40% 🔹 84% of prop traders still fail, even with algorithms 🔹 Only 1–3% of day traders are consistently profitable long term 🔹 Black swan events (COVID, 2008 Crash, 2021 GameStop) confuse AI systems – humans still outperform during chaos 🔹 85% of hedge funds use AI, but still keep humans in control So will AI replace day traders? Probably not. The data shows clearly: AI wins in speed, discipline, and pattern recognition Humans win in intuition, adaptability, and black swan survival Hybrid wins long-term – fewer mistakes, better consistency, higher survival rate What this means if you trade: 🔹 Use AI for scanning, execution, and risk control 🔹 Use your brain for strategy and market context 🔹 Build a repeatable edge before automating 🔹 Stop fearing AI, use it to your advantage ⚡What do you think, will AI replace traders in the next 10 years? ----------- This is not financial advice. Markets are risky, and you may lose money. ---------- This analysis is based on data from: 🔹 81,300+ traders 🔹 28.5 million trades 🔹 150+ peer-reviewed financial studies 🔹 45 institutional AI trading reports

Financial firms aren’t just “experimenting” with AI anymore - they’re letting agentic AI systems make autonomous decisions in trading, portfolio optimization, and risk management. Several recent reports point to a clear trend: AI isn’t just assisting humans… it’s beginning to act like a junior portfolio manager. Here’s what the latest industry sources show: 𝗧𝗿𝗮𝗱𝗶𝗻𝗴 𝗗𝗲𝘀𝗸𝘀 𝗔𝗿𝗲 𝗧𝗲𝘀𝘁𝗶𝗻𝗴 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗔𝗜 Finextra reports that banks and funds are now piloting agentic AI that can propose or execute trades, adjust strategies in real time, and dynamically respond to market conditions. These systems can monitor thousands of signals at once — something even the best human trader can’t. 𝗣𝗼𝗿𝘁𝗳𝗼𝗹𝗶𝗼 𝗠𝗮𝗻𝗮𝗴𝗲𝗿𝘀 𝗔𝗿𝗲 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗻𝗴 𝗔𝗻𝗮𝗹𝘆𝘀𝗶𝘀 FTI Consulting notes a shift from “research assistance” to decision automation, where AI models recommend rebalancing, run scenario simulations, and monitor compliance constraints with minimal human touch. 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 “𝗠𝗮𝗿𝗸𝗲𝘁 𝗕𝗼𝘁𝘀” 𝗔𝗿𝗲 𝗘𝗺𝗲𝗿𝗴𝗶𝗻𝗴 Industry analysis from Phacet Labs describes agentic AI agents that can negotiate orders, scan liquidity, and refine strategies continuously - more like autonomous systems, less like traditional algorithms. 𝗕𝗶𝗴 𝗙𝗶𝗿𝗺𝘀 𝗔𝗿𝗲 𝗔𝗹𝗿𝗲𝗮𝗱𝘆 𝗠𝗼𝘃𝗶𝗻𝗴 Traders Magazine highlights how platforms like BlackRock’s Aladdin are incorporating AI agents into portfolio analytics and risk modeling. These systems aren’t experimental toys - they’re entering enterprise production. ⚖️ 𝗪𝗵𝗮𝘁 𝗧𝗵𝗶𝘀 𝗠𝗲𝗮𝗻𝘀 𝗳𝗼𝗿 𝗟𝗲𝗴𝗮𝗹 𝗧𝗲𝗮𝗺𝘀 𝗖𝗼𝗺𝗽𝗹𝗶𝗮𝗻𝗰𝗲 & 𝗥𝗶𝘀𝗸 ** SEC/FINRA will expect model validation, auditability, and clear supervisory structures. ** Firms must show that AI behavior aligns with the disclosed investment strategy. ** Automated decision-making raises fiduciary duty questions if models deviate or hallucinate. 𝗟𝗶𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗘𝘅𝗽𝗼𝘀𝘂𝗿𝗲 Who is responsible if an AI agent executes a harmful trade? ++ Are errors “system failures,” “algorithmic misconduct,” or “supervision gaps”? ++ Standard E&O policies may not cover autonomous decision-making. 𝗖𝗼𝗻𝘁𝗿𝗮𝗰𝘁𝗶𝗻𝗴 𝗡𝗲𝗲𝗱𝘀 Firms will need provisions for: -- Model-change notifications -- Audit rights for training/data pipelines -- Guardrails on agent autonomy -- Traceability logs for every automated action -- Indemnities for AI-generated errors or unauthorized trades 🔍 𝗚𝗼𝘃𝗲𝗿𝗻𝗮𝗻𝗰𝗲 𝗜𝘀 𝗡𝗼𝗻-𝗢𝗽𝘁𝗶𝗼𝗻𝗮𝗹 As agentic AI shifts from advisory to operational, regulators will expect transparent oversight, documented controls, and explainability - even if the model itself “thinks” in ways humans don’t. 𝗕𝗼𝘁𝘁𝗼𝗺 𝗟𝗶𝗻𝗲 Agentic AI is no longer theoretical. It’s already entering the financial system - and the legal infrastructure around it is years behind. Firms that move now on governance, supervision, and contracting will have a real advantage when regulators inevitably move in.

It’s not just about AI. A wave of converging technologies is quietly rewriting the rules of financial services. 𝟭. 𝗔𝗜 𝗿𝗲𝘀𝗵𝗮𝗽𝗶𝗻𝗴 𝘁𝗵𝗲 𝗺𝗶𝗱𝗱𝗹𝗲 𝗼𝗳𝗳𝗶𝗰𝗲 Machine learning and GenAI are starting to handle tasks that used to require expert judgment: risk monitoring, fraud detection, client communication, and even drafting internal reports. With AI agents in the mix, those tasks become continuous and increasingly automated. Implications: Teams shift from doing the work to supervising it. Accountability now includes both people and systems. 𝟮. 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗔𝗜 + 𝗕𝗹𝗼𝗰𝗸𝗰𝗵𝗮𝗶𝗻 AI agents are moving from tools to actors - able to interpret rules, initiate transactions, and negotiate outcomes. Paired with blockchain’s programmability and auditability, they could create a new execution layer where processes like trade settlement, claims handling, and compliance run end-to-end without human initiation. Implications: Institutions move from managing processes to governing autonomous systems - with real legal and risk consequences. 𝟯. 𝗜𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 𝗮𝘀 𝗮 𝘀𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗰 𝗽𝗹𝗮𝘆 Edge computing, serverless platforms, and blockchain are removing old bottlenecks. Banks no longer need to build everything themselves - they can plug into modular services and orchestrate ecosystems. That opens the door to faster launches, more partnerships, and entirely new business models. Implications: Agility becomes a competitive edge. Legacy stacks risk becoming liabilities, no matter the brand strength. 𝟰. 𝗧𝗵𝗲 𝗨𝗫 𝘀𝗵𝗶𝗳𝘁 With the rise of conversational AI, augmented reality, and natural language interfaces, users won’t just tap and scroll. They'll talk, ask, simulate, and explore. These experiences will reshape how people make financial decisions and who they trust to guide them. Implications: UX becomes a trust driver. How people experience your services will define whether they stick around. 𝟱. 𝗣𝗿𝗼𝗴𝗿𝗮𝗺𝗺𝗮𝗯𝗹𝗲 𝗺𝗼𝗻𝗲𝘆 Attention is shifting from speculative crypto to practical digital money - stablecoins, tokenized deposits, and programmable rails that regulators are beginning to legitimize. They won’t replace traditional finance, but they will build faster, more flexible rails alongside it. Implications: Banks and central banks must engage with new rails - or risk losing influence over them entirely. 𝟲. 𝗣𝗿𝗲𝗽𝗮𝗿𝗶𝗻𝗴 𝗳𝗼𝗿 𝗾𝘂𝗮𝗻𝘁𝘂𝗺  Quantum computing is a long-term development, but its security risks are already relevant. As the pace of advancement accelerates, safeguarding critical systems today is a necessary investment in future resilience. Implications: Security strategies will start to shift from defence to preparation. Opinions: my own, Graphic source: KPMG 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐛𝐞 𝐭𝐨 𝐦𝐲 𝐧𝐞𝐰𝐬𝐥𝐞𝐭𝐭𝐞𝐫: https://lnkd.in/dkqhnxdg

I think there are three disruptive forces reshaping Wall St 1. Fintech APIs 2. Fintech AIs 3. Tokenization They apply across all products and services, and there’s a patchwork of companies and founders attacking them. (Pictured: A simplified mental map) Capital Markets are: 👉 Highly regulated 👉 Highly manual 👉 Hard to upgrade 👉 Full of sunk costs But that’s changing. But there are three big shifts occurring a) Banks have played less of a role since 2008; meaning new actors can fill the gaps b) Alternatives assets are becoming popular but lack market structure like exchanges or CSDs c) We haven’t had a major tech upgrade since the 1970s and are due one Where are the opportunities? 1⃣ Fintech: BaaS for markets? API-first companies are already here. Companies like Drivewealth and Atomic Invest have collectively brought securities and some alternatives to market with APIs 👉 This will go wider into alternatives next 2⃣ Fintech: Platforms for debt raising. Companies like Finley, VaaS, and Setpoint help manage a debt facility. The Arc’s* venture debt marketplace is also an interesting twist on this idea. 👉 Every treasury team will have financial markets embedded in their treasury management software 3⃣ Fintech: Next Gen trading platforms and broker dealers. Younger funds and fund managers will likely adopt the lowest friction UI for trading. Architect* is an example of a company that offers derivatives trading to pro-sumer, startup, or mature funds looking for an ultra-high-performance execution platform. 👉 The next Apollo or Blackstone likely already exists and is currently quite small but using new tools (e.g. @LumidaWealth ) 4⃣ AI: AI analysts and agents. Agent Smyth analyzes stock and sector data to help traders prepare for their day or make a buy/sell decision. Lucite provides companies' business overviews, competitive analysis, and financial metrics as an analyst might export from Pitchbook. Finster is a former deepmind and JP Morgan team-building financial data analysts 👉 Adopters will be a new generation of funds entering capital markets or in their early days 5⃣ AI: AI workflow tools. PDFs are the new oil. This model is popping up all over in Fintech, but it’s especially applicable to capital markets. Everything in capital markets runs on a PDF or spreadsheet, from KYB to ISDA master agreements. 👉 This is the low-hanging fruit use case and makes financial markets products embeddable and 10x less friction 6⃣ Tokenization: Tokenization of money market funds allows 24/7 access. Blackrock has launched a tokenized money market fund BlackRock’s new BUIDL fund, a month after its launch, managing $304 million in assets. Why? 👉 Every CFO wants instantly liquid, high yield products. That isn't true for MMFs today but is with tokenized funds 7⃣ Tokenization: The tokenization of all assets is next. Cash, Stablecoins, Private Credit, Private Equity and commodities are all trading trillions of notional as tokens already

Agents, AI and automation in financial and commodities markets. Why the real revolution is happening in chat. For decades, some of the most valuable data in financial markets has been hiding in plain sight. Not in market feeds. Not in order books. But in chat. Every day, trillions of dollars’ worth of intent, orders, axes, market colour, and trade details flow through human-to-human conversations - and then vanish. Historically, this “digital exhaust” was impossible to capture, structure or use at scale. That’s now changing. Fast. A powerful convergence of open chat APIs and AI maturity is transforming chat from a basic messaging tool into an interactive surface for workflow automation, data connectivity, and trade lifecycle orchestration. And the best part? This isn’t theory - it’s happening right now at the most sophisticated trading firms, brokers and platforms in the world. The shift: from Unstructured Chat Data  ➝ Information ➝ Automation ➝ Control The firms leading this transformation follow a clear path: Information – Capture raw, unstructured chat and turn it into machine-readable data. Automation – Use that structured data to eliminate copy-paste, speed up chat based workflows like RFQs, and reduce operational risk. Control – Build in compliance, permissioning, audit, and human-in-the-loop approvals from day one. As Chris Hudson at FIS puts it: “The demand for data is only growing. What used to live in chat is now powering platforms and AI models.” From Bots to Agents – And Beyond... We’re moving from deterministic bots that fetch data on command, to agentic workflows that understand intent, act across systems, and orchestrate entire workflows – directly from chat. -Consolidate fragmented data from dozens of rooms into a single blotter. -Trigger pre-trade checks and risk calculations without ever leaving chat. -Surface hidden opportunities across desks in real time. This isn’t about replacing humans. It’s about augmenting them - freeing traders and sales teams to focus on relationships, strategy, and insight while the mundane takes care of itself. The Future: Networked Conversations The next frontier is agent-to-agent interaction - workflows where machines negotiate, match, and coordinate within strict guardrails. For markets, that means faster execution, deeper liquidity discovery, and a new layer of intelligence on top of every conversation. The takeaway is simple: 👉 Chat is not just for chatting. 👉 It’s the foundation of a new automation layer for capital markets. 👉 And if you’re not building around it, you’re already behind. At ipushpull, we’re proud to be at the centre of this shift - powering the connective tissue that turns chat into structured data, structured data into automation, and automation into competitive advantage. The question isn’t if this will reshape trading desks. It’s how fast. Read the article in the comments to hear about how ipushpull customers create value from enterprise chat.

Wall Street spent decades hiring people who are good at Excel. That skill just became a commodity. Anthropic just embedded a plugin directly into Excel. This isn't just another add-in. It's a fundamental workflow compression that turns hours of analyst grind into minutes. Here’s my prediction for how this disrupts investment teams: → Modeling & comps get 2-5x faster. → Junior analyst work (formatting, linking) is automated. → Point-solution vendors get marginalized. → Teams move from "build model, write note" to prompted playbooks. → Firms with internal prompt libraries will dominate. → Mechanics commoditized. Firms' edge is better data + prompts. Most firms think this is a productivity play. It's not. It's a total workflow restructuring. The new game is interpretation, not construction. Proprietary prompts, not modeling speed. Unique insight, not standard output. What you should do today: → Pick 3 core workflows (earnings updates, comps) to industrialize. → Start building your internal prompt library and evals now. → Stand up governance. Whitelist your data connectors. AI agents are coming for the core finance stack. It’s a new ball game, learn to play it. P.S. Save this post for future reference. P.P.S. I write a weekly newsletter on how AI is changing investing.  https://lnkd.in/gXwgzJeJ

Retail trading is no longer random—it’s increasingly automated, clustered, and rhythmic. This chart shows SPX 0DTE option volume per minute (non-institutional) for 2025 YTD. Notice the sharp, predictable spikes in activity at specific timestamps throughout the day: 10:00, 10:15, 11:00, 13:00, 14:00… and a surge into the close. - These patterns aren’t driven by humans—they’re automated trade flows, often triggered by preset volatility levels, delta targets, or gamma thresholds. - Algorithms are optimizing execution timing, minimizing impact while maximizing optionality. - Retail platforms are increasingly offering plug-and-play strategies, enabling users to engage in institutional-style intraday scalping without manual input. What used to be a “casino” is becoming a coded battlefield—with latency, precision, and time slices defining edge. The rise of 0DTE is reshaping not just risk-taking—but the very rhythm of the market. Graph Source: Cboe Global Markets

Reimagining Trade: Making a Legacy Industry Exciting Again Trade has long been viewed as a traditional, slow-moving sector dominated by paperwork and old-school practices. Yet, technology is set to shake up this image, turning trade into a vibrant field for innovation and career opportunities: I would never have thought that 20 years ago I would be innovating in trade today. My view of trade was boring old people and lots of paper..In fact, it was short of a “dinosaur industry” in my earlier view. Legacy Challenges: - Manual processes and lack of transparency have traditionally bogged down trade. - Slow adoption of new technologies has kept the industry from evolving quickly. Tech Revolution: 1. Blockchain/Distributed Ledger (yes still a good option) : Enhances transaction security, cuts down on fraud, and speeds up processes with smart contracts. 2. AI and ML: Predicts logistics patterns, automates inventory, and provides insights for better decision-making. 3. IoT: Real-time tracking of goods, improving supply chain visibility and product quality control. 4. Digital Platforms: Connects SMEs globally, automates trade documentation. 5. Automation: Increases efficiency in warehouses and ports with robotics. New Career Paths: - Tech Integrators for system improvements. - Data Scientists for trade analytics. - Cybersecurity Experts to protect digital trade data. - Supply Chain Innovators to optimise global operations. Why Trade is Now Cool: - Innovation Hub: Ideal for tech startups focusing on logistics solutions. - Global Influence: Tradetech impacts worldwide economics, sustainability, and ethics. - Dynamism: A field where technology meets global commerce, offering both challenge and impact. In summary, trade is shedding its 'old and boring' image, becoming a dynamic arena for tech enthusiasts and entrepreneurs. It's a sector where one can innovate, influence global markets, and build a career at the intersection of technology and international business.

AI Is Finally Coming to Trade Compliance And It’s Going to Reshape SAP GTS & Oracle GTM Faster Than Anyone Realises. Over the last few months, I’ve noticed a clear shift in the conversations I’m having with SAP GTS and Oracle GTM/OTM leaders across the U.S. A quiet one, but a big one: AI is finally entering the global trade compliance space properly. Not theory. Not hype. Actual capability. And it’s going to change the entire talent landscape. Here’s the reality no one wants to say publicly yet: The old, manual, spreadsheet-heavy version of trade compliance will not survive the next 3–5 years. Not when AI can classify products, read documents, predict duty exposure, and flag compliance risks in seconds. But here’s the more important point, the one that matters to CEOs, VPs of Trade, and Global Trade Directors: AI isn’t replacing trade compliance teams. It’s exposing the gaps, the bottlenecks, and the lack of system maturity. Companies running legacy GTS 11… Teams relying on manual ECCN/HS classification… Organisations still doing origin audits by hand… All of these are becoming risk points. We’re already seeing early AI use cases appear inside GTS and GTM: ML-assisted HS/ECCN classification AI-driven restricted-party screening noise reduction Document intelligence reading COOs, invoices, packing lists Predictive duty forecasting and tariff-shock modelling Copilot-style assistants for GTS/GTM workflows This is the beginning of the next era of compliance systems the jump from reactive to predictive. And here’s the kicker: The companies that get ahead of this shift won’t just reduce risk. They’ll gain competitive advantage. They’ll clear shipments faster. They’ll minimise duties earlier. They’ll automate what used to take teams of analysts. They’ll operate with real-time visibility, not month-end panic. And yes, they’ll need a new kind of talent. Not admin processors. Not super users who know a few screens. But hybrid analysts who understand: SAP GTS or Oracle GTM Data AI-driven workflows Risk Business impact IMHO This is where the market is heading quickly. I’ll be breaking down each major AI use case across the week. But for today, here’s the simple message: AI won’t replace trade compliance. But trade compliance professionals who refuse to adopt AI absolutely will get replaced. The smart companies, the ones hiring right now, already know this. Would be keen to get your thoughts on this one #GlobalTradeCompliance, #SAPGTS, #OracleGTM

AI trading isn't just replacing humans. It's amplifying our decision-making power. Here's why combining AI & human intuition is the future of finance: 1. Pattern recognition on steroids AI analyzes vast datasets in seconds, spotting trends we'd miss. 2. Emotion-free execution (same with algo trading) Algorithms don't panic sell or FOMO buy. They stick to the plan. 3. Real-time adaptability ML models adjust to market changes instantly, 24/7. 4. Sentiment analysis at scale AI gauges market mood from millions of data points. 5. Risk management precision Models recalibrate risk constantly, protecting your portfolio. But here's the catch: AI isn't perfect. It can't account for black swan events or sudden policy shifts. That's where human expertise comes in. We provide context, creativity, and critical thinking that AI (still) lacks. The future belongs to those who can: 1. Understand AI's strengths and limitations 2. Leverage ML insights effectively 3. Apply human judgment to refine AI outputs It's not man vs. machine. It's man + machine vs. the market.