Great work Chris. This ranking should alarm every Australian and every policymaker. @AlboMP ’s CGT changes have made the family home the most tax-efficient asset in Australia and productive investment the least. That’s not reform. That’s how you lock capital in houses and starve growing companies of the money they need. A country that taxes aspiration will get less of it. PS LICs would go with ETFs in category B.

Quantum Momentum Quantum startup funding rose 6.3x last year, Washington is taking equity stakes, Google's Quantum Echoes advantage claim still stands and AI is a tailwind. Our latest note explores the breakthroughs and the challenges ahead 👇 alphatarget.com/insights/qua…

AlphaTarget’s Insights page publishes a new free thematic research note every month, with our entire archive available to explore. Our thematic research notes cover the big trends shaping technology. No subscription required. Explore below 👇 alphatarget.com/insights/

Aug OpenRouter: open-weight took most of the token volume. That’s good enough+cheaper taking share. OpenAI just made GPT-6 workhorse models a lot cheaper, which is a play for the volume layer. Intelligence is becoming cheaper, fast. A win for adoption & the picks and shovels.

Trend Update $NDX $QQQ in a trading range, the trend is UP NASDAQ-100 holding steady despite all the bad news and we're about to enter the most profitable phase of the US Presidential Cycle. Our portfolio is positioned for the opportunity. alphatarget.com

FSD Supervised safety results from the last 12 months

Celebrating one year of FSD Supervised in Australia & New Zealand with more than 160 million km That’s about 4,000 laps around Earth

INCOMING FSD 14.3.9 Down Under!!! Including summer update! Downloading over cellular while my car parked at the airport ready for my drive home tonight ❤️ Tesla!!!

Right. Dario was already claiming that GPT2 was too dangerous to open source back in 2019. I made fun of them then. Everyone should make fun of them now.

Arriving early doesn't necessarily mean winning!

AlphaTarget is now more than two years old, and I’ve been reflecting on what we’ve learned. We’ve beaten the S&P 500 by a wide margin. I’m proud of that but what matters even more to us is how we build on it from here. The last two years have been an extraordinary period for AI and technology and they’ve taught us a great deal about both opportunity and risk. We’ve now become much more deliberate about portfolio construction and risk management. We still want to own the companies with the potential to reshape industries and compound capital for years but we also recognise that growth stocks have become super volatile and periods of extreme optimism are inevitably followed by sharp resets. That means more active portfolio management, taking gains when appropriate, genuine hedging when conditions warrant it and holding more cash when markets become excessively extended. We refined our portfolio management and hedging strategy this spring, and June and July provided the first serious test. During those two months, many AI stocks fell 40–60%. AlphaTarget’s portfolio declined roughly 20%. It was still a testing period but the portfolio held up considerably better than it had during previous periods of market volatility. We don’t view that as the end of the story, it’s the beginning of a better process. Our objective for the next several years is straightforward: Find exceptional companies early. Participate aggressively when the opportunity is greatest and manage the portfolio actively to reduce drawdowns when the market inevitably becomes volatile. The opportunity created by AI and the technologies surrounding it remains enormous. At the same time, the longer this bull market runs, the greater the risk that parts of the market become speculative (real risk of an AI bubble in the future). I believe navigating that environment - knowing when to be aggressive and when to protect capital - will be one of the defining investment challenges over the next 2-3 years. And frankly, I think the next 2-3 years could be even more interesting than the last two. To all our subscribers, thank you for being on this journey with us. If you’re discovering AlphaTarget for the first time, our research service provides institutional-quality research on the most promising companies in the public markets, along with timely Trading Alerts so subscribers can follow our portfolio decisions as they happen. alphatarget.com

Some of the “leaders” & “contenders” are extinct. 😛😂

Even solana:J3NKxxXZcnNiMjKw9hYb2K4LUxgwB6t1FtPtQVsv3KFr is actively managed. “Buy and hold forever” is a myth. Over the past decade, 200+ companies have been removed from the S&P 500 Index and a similar number added. A 2023 study found ~180 stocks have been replaced since 2015 - about one-third of the index. Of the 2010 roster, ~185 tickers were gone by 2026. The index is a "living" basket. It promotes what is working and demotes what has broken. That is active management with a committee and a rulebook. The same lesson applies to individual “great stocks.” $PYPL was a generational compounder until July 2021 at ~$306. It then lost more than 80% of its value. $LULU printed $511 at the end of 2023; it is now around $122 (~80% decline). $TTD ran to ~$140 in December 2024, then suffered an ~90% drawdown. $NKE and $BA remained household names and index members while years of gains vanished. $ENPH, $WBA, $PAYC and $MTCH were later removed after the damage was already done. That is the point most buy-and-hold sermons skip. Quality is not a substitute for timing. Leadership decays. Growth slows. Multiples compress. A stock can be an excellent business and still destroy capital if you own it through a Stage 4 decline. The index itself does not “hold forever.” It sells the laggards - often late, but it sells them. At AlphaTarget we treat all this with discipline. We combine fundamental research with technical analysis to identify the leading disruptive growth stocks, then use Stage Analysis to time entries and exits: accumulate in emerging Stage 2 uptrends, ride the trend while the tape confirms, and step aside when the structure breaks. The aim is not to marry a ticker. It is to own leadership while it is leadership - so we can hold onto the years of gains that disappear in months. The market does not pay you for loyalty - it pays you for being in the right stocks, at the right time. Even solana:J3NKxxXZcnNiMjKw9hYb2K4LUxgwB6t1FtPtQVsv3KFr admits this; Individual investors should too. alphatarget.com

For the first time ever, EVs outsold ICE vehicles in Australia 🇦🇺 Model Y was the #1 vehicle & Tesla the #3 best-selling brand overall for the month of August drive.com.au/news/australian…

Strong results from $DELL and a BIG $25B raise to their FY27 revenue outlook. FY27 revenue now expected to grow 69% to $192B. AI is a big driver. Chart: $DELL TTM Revenue & quarterly YoY revenue growth from @ycharts

Taking over from Steve Jobs would have been hard. The bar was sky high. But Tim Cook cooked hard for 15 years. Those shareholders who came along for ride were richly rewarded. Thank you @tim_cook 🙏

PSA The ping on this @viasat powered @Qantas flight from Sydney to Perth is longer than bouncing off the moon! 🙄

Sentiment has become more cautious and the market has switched to "risk off" mode after Warsh’s Jackson Hole speech. The market has gone from treating a September hike as a low-probability scare to treating it as a live coin toss. Financial conditions are not obviously restrictive, inflation has not cleanly rolled over, and the Chair just told everyone the Federal Reserve still has work to do. Until the 15–16 September FOMC is behind us, expect chop, false breaks and a lot of narrative noise. The near-term uncertainty is not the same thing as a broken bull market. The bigger picture is simpler. The June–July washout in the leading AI names was violent enough to look like a regime change. It wasn’t. It was a positioning and technical shock. Those stocks are now attractively valued versus the cycle they are still in. AI metrics are improving - token usage is still accelerating. GPU rental markets remain tight in the parts of the stack that matter. Inference demand is spreading from labs into enterprises. Capex visibility into 2027 has not collapsed. Unless the economy tips into recession, that combination usually precedes a strong Q4–Q1 rally, not a graveyard. A few months ago, software was written off as dead. Today, the high-quality infrastructure software names are back at all-time highs. Even $IGV is only about 7% below its all time high. That is the market telling you the “software is dead” narrative was a false premises, not reality. The same pattern is now setting up in AI. The stocks got marked down in June-July. The usage, pricing power in compute and buildout did not. This is why the next few months matter more than the next few sessions. A hawkish Fed can keep the front end noisy. It does not automatically kill a cycle where real demand for intelligence and infrastructure is still compounding. The opportunity is in separating the noise from the trend. That’s the work we do at AlphaTarget: map policy, positioning, companies’ operating metrics and the actual AI operating metrics - not the headlines - and stay on the side of the secular trend. alphatarget.com

Modern warfare has changed, Western military arsenals haven’t. Using million-dollar missiles to tackle cheap drones isn’t sustainable. Our latest note looks at the defence companies modernising Western militaries. Read below 👇 alphatarget.com/insights/def…

Marvell’s multi-year deal with Google (disclosed 19 August) isn’t Marvell eating Broadcom’s lunch. $MRVL $GOOG $AVGO TLDR: ✅Google: multi-sourcing and adding more optimised/specialised content as it scales its TPU deployment ✅Broadcom: core TPUs + rack/cluster networking fabric ✅Marvell: attach silicon around the TPUs (controllers, NICs, memory, near-memory, some inference) ✅More specialised / disaggregated TPUs: higher interconnect content (more AECs for shorter reaches, plus retimers, SerDes, switches/NICs, and optics as distances grow) Under Google’s April 2026 long-term agreement, Broadcom remains the primary partner for the core TPU silicon. That deal also includes a supply-assurance agreement covering networking and other components for Google’s next-generation AI racks, the high-performance switch and fabric silicon (Tomahawk, Jericho, SerDes, etc.) that sits inside those racks and interconnects the TPU clusters. Marvell’s deal is actually complementary. It covers custom products that attach to the TPU ecosystem, so around the TPU die: AI inference accelerators, storage controllers, NICs, memory interface controllers, and near-memory compute. In simple terms, Marvell is supplying a bunch of the supporting silicon around the TPU, the stuff that helps with data movement, memory hierarchy, storage access, and networking at the node level. The agreement runs out to roughly 2033 and comes with a big performance-tied warrant that vests mostly as Google buys the products. Full vesting implies up to ~$120 billion of qualifying revenue over the period (a ceiling, not a backlog). Google gets to exercise those warrants as it spends on Marvell’s gear.