Ok, I went through the recent $ONDS acquisitions of GATE Technologies, Bron Technologies, and Aran Defense. These happened after the Q2 filing, but as of the end of Q2 ONDS had $1.38 billion of cash and short term investments with only $4.1 million in long term debt on the balance sheet. Yes, they lose money today but their cash position more than exceeds their annualized run rate operating losses. I have my own investment thesis formed in my head but I want to hear from the ONDS community. What is the long-term thesis on ONDS and what are you most excited about?

$ONDS 🦅🔎 “The drone now has tethered power, so we’re no longer working around battery changes. Our engineer can fly from the tanker’s deck, without entering the tank.” “Onboard LiDAR lets us pin points of interest to a real-time 3D point cloud and we no longer need cumbersome temporary lighting setups.” The commercial/inspection TAM for drones is still largely untapped.

Oppenheimer’s Israel desk published a note today on Lemonade $LMND, focusing on the implications of digital agents such as Muse and Grok Bot, ahead of the company’s upcoming Investor Day in November. The translation is mine (well, not really, it’s GPT’s), and any errors are my responsibility: Lemonade (LMND) and Digital Agents – Should the Company Open Its Services and Products to New Digital Agents Such as Grok Bot and Muse, or Block Them and Remain Focused on the Human Interface It Has Spent Years Perfecting? Digital Agents – A Blessing or a Curse? Lemonade, like most commercial companies selling products and services today, is facing a new strategic dilemma. On the one hand, recent technological developments in digital agents are creating an entirely new user interface and sales channel. This channel could significantly expand the number of customers these companies can reach and, accordingly, materially increase revenues. On the other hand, it could significantly weaken the direct relationship between the company and the customer, while reducing customer loyalty. The most recent example is the collaboration between META and SHOP, under which the two companies announced that the MUSE agent would be able to make purchases through SHOP’s platform. Customers would no longer need to enter the website themselves; instead, they could simply instruct their digital agent to make the purchase on their behalf. AMZN, by contrast, has chosen to block MUSE, primarily because it is seeking to protect its advertising business, which generates approximately $70 billion in annual revenue. Ultimately, however, both companies are pursuing the same objective: preserving the uniqueness of their platforms and maintaining the direct company-customer relationship, while at the same time integrating agentic capabilities that could significantly increase sales volumes in the coming years. Clearly, there are many business considerations that are specific to each industry and to each company operating within it. Still, the contrast between SHOP and AMZN illustrates very well the dilemma that almost every consumer-facing company will need to confront over the next several years. LMND’s Dilemma LMND’s dilemma is somewhat different from the one facing the e-commerce industry, primarily because a meaningful share of insurance policies are still sold through human agents, depending on the type of insurance and geography. Lemonade, by contrast, operates an almost entirely digital distribution network, with approximately 98% of sales taking place without a human agent. Lemonade’s real competitive advantage lies in its cost structure. Its policy acquisition and onboarding costs are significantly lower than those of traditional competitors, at approximately 4% and continuing to decline, compared with an industry average of roughly 9%. At the same time, the company’s business model incorporates substantially greater operating leverage than that of traditional insurers. This allows LMND to translate its lower cost structure into lower policy pricing. Assuming that insurance is fundamentally a commodity, with limited differentiation in the underlying product relative to competing offerings, lower pricing should ultimately translate into significant revenue growth. The key question for Lemonade is how to enable digital agents to purchase its products. Given its lower pricing, digital agents should naturally be attracted to LMND’s offering. At the same time, the company must find a way to preserve customer loyalty and maintain the direct relationship with the end user. In our view, the company is likely to unveil at its Investor Day on November 27 a strategy that will allow digital agents to purchase Lemonade insurance while preserving customer loyalty and enabling LMND to continue cross-selling and up-selling additional products to those customers. The most rational approach would be to transform AI MAYA, LMND’s existing bot that sells insurance directly to human customers, into an agent that can interact more effectively with third-party digital agents. If a user instructs his or her digital agent to purchase the cheapest suitable insurance policy, that agent could communicate directly with AI MAYA. MAYA would then collect the relevant information about the human user through data-sharing with the user’s agent. LMND could potentially require the customer’s agent to provide extensive information regarding the user’s online behavior, subject of course to privacy and regulatory constraints. This could allow the company to preserve, and potentially even improve, the quality of its underwriting and policy pricing. Once AI MAYA provides a policy quote, it could continue offering bundled insurance products to the digital agent, such as auto plus pet insurance, or auto plus homeowners insurance. The digital agent could then review the policyholder’s existing coverage and determine whether those additional products are relevant. If the agent ultimately recommends that its human user purchase the policy, AI MAYA could continue interacting with that customer over time, selling additional policies, proactively making new offers, and extending the growth trajectory Lemonade has delivered over the past three years. The critical issue is how LMND can leverage its core advantages relative to competitors: lower pricing, better service, and a strong reputation for reliable claims payment. The objective is not merely to grow the number of policies, but to do so while preserving the quality of the insurance book. In our view, the growing use of digital agents will cause customers to become significantly less loyal to their existing insurance providers. Instead, many customers will simply purchase the cheapest and most reliable insurance policy recommended by their agent. The key question is whether LMND can make itself sufficiently accessible to these digital agents and whether, once a customer has purchased a policy, the company will be able to act proactively by offering that customer additional products and maintaining the relationship over many years. Based on the company’s track record, we believe it can. The Competitors’ Dilemma The use of digital agents could materially undermine the core businesses of traditional insurance companies. For decades, traditional insurers have invested heavily in branding. Yet as purchasing decisions increasingly shift from humans to digital agents, branding may become progressively less relevant, while pricing differences between policies become more important. If a digital agent is instructed to purchase the most reliable and lowest-cost insurance available in the market, it will not be impressed by the billions of dollars that insurers have spent building brands designed to influence human consumers. Instead, it will evaluate the options more dispassionately and determine which policy delivers the required coverage at the lowest cost. In such an environment, LMND could enjoy a meaningful structural advantage. If traditional insurers choose to block these digital agents, however, they may ultimately be hurting themselves. A potentially significant new distribution channel is being created in front of them, and by refusing to participate, they risk being excluded from it entirely. Traditional insurers are therefore entering a difficult strategic dilemma. The only viable solution may be to adopt digital-agent integration as quickly as possible while simultaneously adapting the business models on which they have historically relied. There will, of course, remain a place in the market for insurers offering highly personalized, premium service at significantly higher prices, such as Chubb. But most of the insurance market does not operate in that segment. Bottom Line Digital agents, currently led by platforms such as Muse and Grok Bot, and likely soon to be joined by agentic offerings from the other major AI companies, are likely to place substantially greater emphasis on policy pricing than the average human customer does today. These agents will not be influenced by the enormous budgets insurers allocate to marketing and sales. Instead, they will seek the lowest possible price for the most appropriate policy. The insurance companies with the lowest cost structures, the strongest ability to adapt to new technologies, and the easiest access for digital agents are likely to be the biggest beneficiaries of this emerging distribution channel. We believe LMND is likely to unveil its new strategy in this area at its upcoming Investor Day on November 27, and we believe the company is very well positioned to benefit from this new technological development. Oppenheimer Israel

$IRDM / $RKLB: Iridium just put a 1,000x stronger encrypted LEO timing signal into the infrastructure that keeps 5G, power grids, and defense networks synchronized. With Oscilloquartz integrating Iridium PNT as a live backup, those networks can maintain precise time even when GNSS is unavailable. Another strategic product Rocket Lab is acquiring through IRDM.

Ondas' Cyberhawk secures a three-year SSE agreement for electricity network drone inspection and surveying, adding offshore wind inspection. A renewed multi-year iHawk agreement supports risk prioritization and asset management. $ONDS ondas.com/post/ondas-cyberha…

At NFL games this season, drone defense tech aims to bring down disruptions cnbc.com/2026/09/27/nfl-dron…

$OUST traded a total of 53,290 contracts this week, of which 79% were calls.

$LMND has the vp of gen ai / muse product on the board. Meta superintelligence labs. muse.ai is on his linkedin. in the room since october. couldn’t stop thinking about that because muse is a new way people are gonna buy insurance. agent shops it, fills the form, switches the policy. ugly for allstate and progressive websites. those were built for a person clicking around for 40 minutes. didn’t even make this connection until last night and i was like holy shit. if LMND is the choice for meta as they start distributing through muse that’s gonna be insane.

BLUE ORIGIN RAISES $10B AT $140B VALUATION Jeff Bezos has now invested roughly $30B into Blue Origin since founding the company in 2000, including another $2B as part of its first outside funding round. Internal projections show revenue rising from: 2025: ~$800M 2026: ~$1.4B 2030: $30B+ Blue Origin expects growth to come from New Glenn launches, a planned satellite communications business and a future fleet of AI satellites. The company currently employs ~15,000 people and is rebuilding its Florida launch complex after a New Glenn vehicle exploded on the launchpad in May. Source: WSJ

$META LAUNCHES AI GAME-BUILDING TOOLS Meta is launching Horizon Create and Horizon Studio, letting users build complete 2D or 3D games from AI prompts on mobile or in a browser. Games can then be published across Facebook and Instagram, where users can jump from a clip directly into a multiplayer session without downloading a separate app. Meta says the tools can generate gameplay systems, difficulty, art direction and multiplayer features, while still letting creators manually refine the experience. The rollout is part of Meta’s broader shift to make Horizon a mobile-first platform rather than primarily a VR product.