كان لي شرف نشر مقالي بإدارة رئيس تحرير مجلة المحامي الصادرة عن جمعية المحامين الكويتية الموقرة أستاذي الدكتور محمد غالي العنزي @galilawyer تحت عنوان: "الخوارزميات التي تُحرّك أسواق الم��ل.. من التداول عالي التردّد إلى وكيل الذكاء الاصطناعي (هل سيصل المد التقني إلى بورصة الكويت)"
AI may be grabbing the headlines, but the broader technology ecosystem is where the scale becomes staggering— what industry do you think deserves more attention? Infographic via @Eli_Krumova x https://t.co/kyLvacle1g; 4 key insights:
1⃣💻 Technology Is in a League of Its Own: At $51.4 trillion, Technology's market capitalization is more than double the next-largest category, AI ($23.9 trillion). Its valuation reflects the enormous scale of the broader technology ecosystem.
2⃣🤖 AI Is Already Competing With Established Technology Giants: AI ($23.9T) has edged past Tech Hardware ($23.3T), a difference of just $600 billion. AI has reached a scale comparable to one of the world's most established technology industries, illustrating just how significant its market presence has become.
3⃣🏦 Financial Services Is Bigger Than Banking Alone: Financial Services carries a $21.0 trillion market cap, compared with $14.9 trillion for Banks. The broader financial ecosystem extends far beyond traditional banking, encompassing a much larger market of financial businesses and activities.
4⃣🔬 The Technology Economy Extends Far Beyond Software and AI: Semiconductors ($17.8T), Electronics ($16.5T), and Manufacturing ($13.5T) each represent enormous market categories. The technology economy is built on physical infrastructure, components, and industrial production just as much as software and AI.
#DigitalTransformation #FutureOfWork
New. "5 Days, 3 Major Disruptions: What Microsoft, the FAA, and Major Banks Reveal About Contingent Business Interruption Risk" https://t.co/kCBqQ8m1hY
Major outages across critical infrastructure— how does your enterprise prepare for volatility?
📌 Key Points for Leadership
🧑💼 A Week in Dana's Shoes
🗓️ The Week in Brief
🔍 Was This a "Stress Test"?
🛡️ Where Insurance Enters the Story
✅ What to Ask Your Broker
🧭 What Leadership Should Do in the Next 30 Days
🏁 The Takeaway
#RiskManagement #Reslience
Insurance runs on "sticky" customers— AI agents don’t do loyalty. How should carriers compete when the shopper isn’t human? 4 key insights; full article via @felixstocker14 x @foreseeinsure below:
1⃣🔍 Carriers Call Agentic Shopping a “Rounding Error,” But They’re Preparing Anyway: Progressive, Travelers, and Lemonade all say AI-driven traffic is tiny so far, and mostly experimentation. Progressive is still getting ready for it to become a real acquisition channel.
2⃣💸 Sticky Customers Fund the Whole Industry: Insurers earn their margins from customers who overpay and never switch. If AI agents make comparison shopping effortless, that cushion shrinks. People already switch home and auto policies at 2% and 4% per month, and that rate could climb.
3⃣🛡️ Carriers Want to Keep Control of Quoting and Binding: Insurers have spent hundreds of billions building their brands, so they won’t hand the customer relationship to a third-party agent. Every quote also costs them money, and agents skip the ads that carriers pay for.
4⃣🏆 AI Assistants May Favor the Biggest Brands: One recent paper found that dominant brands get cited 1.4x to 2.8x more often by AI assistants. If that holds, big incumbents could keep their advantage even as shopping shifts to agents.
#AgenticAI #InsurTech
Breaking Now: "When the Skies Stop: Business Interruption Risks Behind America’s Aviation System." https://t.co/NX3Ekg7i1J
The FAA just issued a nationwide ground stop— who in your supply chain just stopped moving?
🛫 What We Know
❓ What We Don't Know
🌐 Beyond Aviation: Who Else Absorbs the Loss
🛡️ The Insurance Lesson: Economic Loss Is Not Insured Loss
🏗️ The Infrastructure Warning
✅ What Leaders Should Do Now
#FAA #GroundStop
New via @forus (courtesy @hollyyzhuang). AI agents just turned a days-long healthcare process into an hour-long one. What’s the most broken process in your industry waiting for the same fix? 4 key insights; full article: https://t.co/COgJ4MPlKo
1⃣🔁 The Answer Is Never Final: New card, pharmacy note, or rejected fax? Germinator re-reads the whole file and re-decides. Routing answers expire when the insurance answer changes.
2⃣🧠 ��I Don’t Know Yet” Is a Valid Output: Agents that can abstain, gather evidence, or escalate beat agents forced to guess. Even a “patient not found” reply is useful signal.
3⃣🔒 No Evidence + No Named Rule = No Action: Every decision must cite both, no matter how confident the agent is. The logs also let you replay past cases against new models before trusting them.
4⃣📊 Specific Beats General, and Scale Beats Everything: The patient’s own history outranks plan-wide patterns, which outrank generic assumptions. Pooling outcomes across a network produced 98.1% of routing decisions with no human intervention (per Forus).
#InsurTech #AgenticAI
New via @umanitek (CC: @origin_trail). AI is ushering us into an era where political impersonation is more sophisticated and realistic than ever— how do you decide whether a video of a public figure is real? 4 key insights:
1⃣🎭 A Familiar Face Is the Weapon: Impersonation borrows credibility that already exists. A candidate’s face, voice, and identity can be replicated with AI, so public trust in the real person becomes part of the deception.
2⃣🎬 AI Can Fabricate Events That Never Happened: AI-generated audio, video, and imagery can make a political figure appear to say or do things they never did, and the result can look authentic even when the event is fiction.
3⃣📡 Speed and Scale Favor the Fake: Fabricated content can spread rapidly across social platforms, messaging apps, and online communities. By the time it’s identified, it may have already reached a substantial audience.
4⃣🚧 Detection Is Not the Same as Protection: Spotting manipulated content is only the first step. Effective defense also requires coordinated reporting, removal efforts, and measures to limit further spread.
#Deepfakes #Disinformation
NEW Exclusive: "When the CEO Becomes the Security Risk: Executive Exposure, Foreign Intelligence, & Limits of Insurance." https://t.co/V3ZZxMqTuC
Trust is the one asset no insurer can replace— how should boards protect a company from risks starting at the top, especially during the global AI race?
📩 A Sunday Memo and a Bigger Question
🔎 What Is Confirmed, Alleged and Unknown
⚖️ The Governance Test: Directors, Officers and Process
🕵️ Espionage Without a Firewall: Where Coverage Runs Out
🌐 Trust, Geopolitics and the Founder Premium
🧭 The Central Question: Can You Insure Against Your Own People?
✅ Seven Questions for Your Board and Risk Committee
🎯 The Bottom Line
#NVDA #NVIDIA #JensenHuang
Lemonade Inc insurance cofounder @shai_wininger reporting twelve straight quarters of accelerating growth, and the next customer may be an AI agent— how should insurers prepare for a world where software does the shopping? 4 key insights:
1⃣🚀 Growth Decoupled from Headcount: 2.5x the premium base with fewer employees than four years ago. Automation across onboarding, underwriting, and claims means each new premium dollar costs less to serve.
2⃣🧩 One Platform, Endless Launches: 25 new product-geography combos this year. A reusable core means expansion compounds instead of adding complexity.
3⃣🚗 Data-Driven Car, with Tesla FSD as the Wedge: Car IFP is up 63% across 14 states. Autonomous-driving coverage lets telemetry price risk better than legacy rating factors.
4⃣🤖 Built for the Agent Era: Native personal AI agent support is coming. An API-native, fully automated insurer is easy for agents to quote, buy from, and service.
#InsurTech #AgenticAI
One missed detail in professional services like healthcare, law, insurance, education, finance, etc can cost someone everything— AI is already touching all of them. Who is making sure your industry knows how to use it responsibly? 4 key insights:
1⃣🩺 Google Built A 5 CME Credit AI Course For Clinicians: Physicians at Google created a self-paced certificate approved for up to 5.0 AMA PRA Category 1 Credits. The whole thing takes under 5 hours.
2⃣⏱️ Reclaim Time From Paperwork, Not Patients: Learn to pull action items from emails, solve scheduling puzzles, and turn clinical documents into audio briefings. The goal is more eye contact with patients and less time typing.
3⃣🔒 Privacy And Clinical Judgment Come First: The course teaches you to keep Protected Health Information out of AI tools, to follow your institution’s policy as the final word, and to supervise every AI output yourself. It deliberately does not cover diagnosis or treatment decisions.
4⃣💡 No Experience Needed And Surprisingly Affordable: There are no prerequisites, and the four courses run about an hour each. It costs $49 a month on Coursera in the U.S. and Canada, so finishing fast keeps the price low.
#HealthTech #FutureOfWork
AI just turned a 120-page health insurance PDF into a one line answer�� what’s the first thing you’d want to know about your own plan? 4 key insights on putting LLMs to work reading your policy:
1⃣🔄 Turning Unstructured Documents into Structured Data: Insurance policies are long, dense PDFs that people rarely read in full. LLMs can parse a 120-page coverage document and extract the details (copays, deductibles, coverage rules) into a clean, usable format. This is the core capability that makes everything else possible.
2⃣💵 Turning Policy Language into a Real Price: Once the data is structured, the answer becomes a concrete number, like “your copay is $30,” rather than legalese. LLMs bridge the gap between what the contract says and what you’ll actually pay.
3⃣🪄 Hiding the Complexity: The heavy lifting happens behind the scenes, and the customer just sees a simple answer. The best use of an LLM here is often invisible; it does the translation work so the experience feels as easy as any other consumer product.
4⃣🔍 Creating Price Transparency That Didn’t Exist Before: As the post points out, there are very few places in healthcare where you can see a price before you buy. By making policy details readable and queryable at scale, LLMs make it possible to know your costs upfront, which was previously impractical.
#HealthTech #ConsumerHealth
History made this week: the first NFL cheerleader labor union has been officially recognized. https://t.co/nn7KchWMnR
Read about past labor disputes and much more. What could be next for the DCCA (Dallas Cowboys Cheerleaders Association)?
#NFL#EmployeeBenefits
New. "Dallas Cowboys Cheerleaders Union: What Employers Need to Know About Pay, Benefits and Risk." https://t.co/gZ6NuYY4ij
The Dallas Cowboys cheerleaders already got a huge WIN this week: unionization. What could be next for their pay, benefits and risk?
✅ What Is Confirmed So Far
⚖️ A Decade of Wage Disputes Sets the Backdrop
🏷️ Employee or Contractor? Classification Drives Everything Else
🩹 Workers' Compensation: The Injury Question, and a Texas Twist
🛡️ Employment Practices Liability: Where Wage Claims and EPLI Meet
🤝 Benefits at the Bargaining Table: What It Might Look Like
📋 What Business Leaders Should Take from This
🧭 Where Advisory Support Fits
#NFL #EmployeeBenefits #RiskManagement
Scaling & reinvention through agentic AI. @PodiumHQ (via @ericwilliamrea) reached $150M in AI ARR in 2.5 years, a third of the time it took the first go-around. With 20K agents working for customers, they’ve stopped selling tools and started delivering results— what would your company deliver if agents did the work?
1⃣🔄 Disrupt Yourself Before Someone Else Does: Podium had a $100M+ software business and rebuilt around AI anyway. The second $150M took 2.5 years instead of 8.
2⃣🤖 Sell the Work, Not the Software: Podium’s 20,000 AI agents do the work for customers and have helped generate over $13B in revenue. Value is measured in outcomes, not features.
3⃣🏗️ Own the Whole Stack, From Agent to System of Record: New industry operating systems put agents on top of the data layer. Customers now run their entire business on it, which makes it far stickier than a point solution.
4⃣⚡ The Product Is Only Half the Reinvention: Podium also rethought how it builds, sells, and serves, essentially a second startup inside the first. It did all of this while staying profitable and cash-flow positive.
#AgenticAI #EnterpriseAI
6G connectivity aims to make the physical and digital worlds indistinguishable in real time— what happens when AI is built into the network itself, not just running on top of it? 4 key insights; visit https://t.co/nNUafKsZej to follow along:
1⃣🧠 Networks That Learn, Not Just Carry: With AI inside the network, it can optimize, heal, and adapt itself in real time instead of waiting for engineers to configure it. The network shifts from a static pipe to a living system.
2⃣📡 Sensing Becomes Understanding: When radio signals double as sensors, AI can interpret what they reveal about movement, environments, and context, not just move data. Connectivity becomes awareness.
3⃣🌱 Intelligence Makes Efficiency Possible: AI can predict demand and power down idle capacity, so more performance doesn’t have to mean more energy. Smarter networks can also be greener ones.
4⃣🔐 Trust Has To Be Built In: If AI makes decisions inside the network, transparency and reliability become part of the infrastructure itself, not an afterthought. The more the network decides, the more trust matters.
#FutureOfTech #Innovation