Here’s a preview of what we’ll cover this week: 

Macro: How’s the Economy Doing?; What Will Warsh Do?

Markets: Three Reasons For A Bounce; Meta’s Guide To AI Capex; Should Anthropic Be Priced Like Meta?; OpenAI Is Renting Distribution

Lumida Curations: Elon Musk: Superintelligence Alone Isn’t Transformative; Ken Griffin: GPUs Have an Airline Problem; Dylan Field (Figma): AI Lowers the Floor, Not the Ceiling

Our conversation with Dan Madden, Principal at Squadra Ventures, is now available on YouTube. We recorded it at a Lumida Penthouse Socials event in New York.

Dan is a Marine veteran who served in the Office of the Secretary of Defense, and on the House Armed Services Committee, and was previously at RAND. He was also one of the first VCs to focus on defense tech.

Here’s what we covered:

Watch the full conversation here. For more conversations like this one, subscribe to our YouTube channel.

You can also download our defense tech primer here. It covers how AI is rebuilding defense, ten private companies that are part of the shift, and the upside and risks for investors.

AI is now at a tipping point across a number of non-coding domains. 

AI started with coders. But, now it is in personal assistant markets and even in pop culture.

Give this tune a listen on Suno, and tell me AI isn’t at par with humans at music, if not better. 

It still feels a bit unnatural. 

The quality of the music is enough to create demand pull like Pandora back in the day.

We are seeing AI focus on specific verticals now:

You can see it across domains - Harvey (Legal), Rogo (Financial Analysis), Basis (CPA work), and Lumida (Wealth Management).

We are still very, very early in AI adoption. That’s my main takeaway from speaking with VCs and Entrepreneurs at Money 2020. That’s my view from asking them what they see out in the market and the infancy of AI products.

I can see that in our own tooling.

The first clear win with agentic wealth management we see is the discovery.

A human advisor walks into a first meeting with whatever prep time the calendar allowed. Maybe a quick LinkedIn scan, maybe a glance at the intake form.

An agent arrives having read everything publicly available. It has already run the Google search, reviewed the LinkedIn profile and built the picture before the call starts.

It knows who you are before you show up.

On discovery, the agents are already better than humans. Their context base is bigger, and they never skip the prep.

That changes what a first meeting is.

Today, the first hour goes to questions the advisor could have answered alone. Where do you work? What do you own? Do you have a trust?

It’s the most expensive hour in the relationship and the least valuable.

With an agent, the conversation starts where it gets interesting: “You sold your business last year and have two kids under ten. Let’s talk about what that means for your estate plan.”

This is Experience as a Service in practice.

The next-gen winners will compete on experience, not product. Read more about our Experience as a Service thesis here.

We have recently launched Iris, the world’s first agentic advisor. Iris joins you live on a Zoom call and you can ask her questions the way you would with a human advisor. We are open to demos. You can join the waitlist here.

I recently tested it at the Money 2020 event, and the reception was strong. AI agents are coming for service businesses. See some of the anonymized reviews below.

(Disclosure: Don’t rely on either agent for tax or investing advice, and using these agentic services does not establish a service relationship between Lumida and you. We are showing you a glimpse of the future.)

Had a very sharp disagreement with a thoughtful VC. 

Her point was that brands don’t matter, or will matter far less, in the age of AI. 

Her argument:

(1) New brands are being built overnight. (See Higgsfield, Vercel, Supabase, etc.). 

(2) AI reduces the search and discovery layer. 

(3) AI finds the best solution and implements. 

I agree with those two premises but hard disagree on conclusion. 

Yes, we will see new brands that disrupt the legacy incumbents. 

Yes, AEO (brand discovery via an LLM rather than search) is real. 

That’s why GPT is getting into AEO lead gen. 

But… business moves at the speed of trust. 

Reputations will continue to matter. 

They will matter even more in the age of AI when every meeting prep contains a full profile write-up. 

Also, brands are condensed information kernels. 

A brand contains a set of expectations. 

Humans simplify. 

The AI human handlers also will rely on brands. 

Brands will be built up and destroyed faster in the age of AI. 

Brands will matter even more as we navigate a future of startups which are essentially AI slop businesses.

This is the existential question VCs and leaders are asking.

Technical velocity doesn't matter. 

So, what does?

(1) Regulation is a moat.

This is a big one.

Banks and insurance companies are not going to go away. 

The balance sheet nature of these business models will not change.

Licensed physicians aren't going away either. 

So, if you're running an agentic business model in this category against slow incumbents with large TAMs, you can win by taking share from the dinosaur.

That's my thesis with Lumida Wealth.

This is why Palmer Luckey is focused on regulated categories like Defense Tech and Banking Tech.

(2) GTM Velocity beats Technical velocity

How fast can you spin up and test?

Now, this is a "skill" not a moat. 

But if you run fast enough, it looks like a moat. 

Excellence deters a new entry. 

Higgsfield has 150 content creators. That's why they are growing revenue. It's not the product. 

So, GTM velocity looks like a moat. 

(3) Intellectual Property is a moat.

Influencers are a moat. 

Any kind of IP is a moat (Air Jordans, Disney characters, Seinfeld re-runs, pharmaceuticals, etc.)

(4) Founder Agency

Leadership is a moat.

This is an uncomfortable truth. 

Elon Musk is his own moat. 

Elon can marshal capital and talent density on demand. 

That's a competitive advantage that allows him to enter markets with only 1 or 2 competitors (TSMC, Iridium, etc.). 

That's equivalent to a moat.

Zuckerberg is his own moat.

He made the non-consensus call to hire Alexandr Wang. 

No way Alexandr would have survived a traditional 'check the box' HR process. 

Steve Jobs was his own moat. 

His vision and product instincts were unmatched. He could make hard choices and simplify.

Zuck has exceptional talent acquisition instincts.

AI highly levers the Founder Agency.

AI turns Founders into 3x levered beta products.

So, the bets VCs make on Founder talent and leadership matter much more in the age of AI.

(5) Brands are a moat.

Brands are a type of IP. So, brands are a moat.

Some VCs will say 'show me your process and I will create your brand'.

Nope. Wendy's, McDonald's, and Burger King have all entered Chick-fil-A's market with nearly identical chicken sandwich products.

Chick-fil-A still dominates.

(6) Community is a moat.

Sports teams are moats. Political parties are moats.

So are alumni organizations. 

If you can build and compound a community you have a moat. 

(7) Customer intimacy is a moat.

The startup that understands its customer / ICP better than the other is a moat.

The mission is to create habit loops with your customer. 

That means meeting the customer where they are, in their channel of choice, and on their terms.

Even if it means engaging with them with a digital avatar in their pajamas at 2am.

If you're a founder, you should be obsessed with all of these moats.

And, for most of 2026, the macro section could have been written by a San Diego weatherman.

Sunny. 72. Same as yesterday.

We’d love to bring something different this week. The data won’t let us.

The economy is hot.

The Atlanta GDPNow model has Q3 real GDP growing at 3.5% annualized. 

Consumer spending is on track to grow 3.8%, while business investment is growing by more than 11%.

Overall, real final sales across the economy are growing by 4.7% YoY, the fastest pace since Q1 2023. 

We saw the same consumer strength in this week’s earning reports. Here are some insights from Capital One’s earnings. (We own it)

Capital One saw purchase volumes grow 11% from a year ago — that’s the confident consumers in practice. 

The card charge-off rate fell to 3.98% in Q3, down 65 bps from a year ago. Delinquencies came in at 3.74%, also lower year-on-year.

Richard Fairbank (Chair and CEO, Capital One) said the consumers are “in a very good place with respect to credit”.

Recoveries are also improving. Fairbank said, “we see stronger recovery rates on recent charge-offs than what we’ve observed historically.”

Consumers are strong - they are spending, and are paying back on their debts.

What keeps the consumer paying is a paycheck, and the paychecks are still coming.

Overall, we see no cracks in consumers, businesses, or the economy. 

We do see a lot of fear around midterms, geopolitics, and high gas prices.

It was a rough month for tech. The Nasdaq 100 is down about 5% over the past month, with XLK down about 7%. 

The Nasdaq 100 now trades at 20.5x forward earnings, its lowest multiple since the AI cycle began in 2023. 

Its premium to the S&P 500 is the narrowest in nearly a decade.

You’re paying the smallest premium in ten years for the fastest-growing part of the market. That spread won’t stay forever. 

We see three reasons why markets can get a near-term bounce. 

The first is the calendar.

The midterms are nine days away. 

In every calendar year following a midterm since 1943, the S&P 500 has finished higher. 

That’s 21 out of 21, with an average return of 20.3%.

That’s a better record than the Patriots with Brady.

Post the last midterm in 2023, the return came in at 26%.

The election clears the uncertainty. And, then the White House spends the pre-election year trying to make the economy look as good as possible.

The second reason arrives before the mid-term: the return of the market’s biggest buyer.

Goldman estimates only 12% of S&P 500 companies are currently in an open buyback window, with another 22% exiting blackout by the end of this week. 

(See the chart below to see the number of companies in an open-buyback window each day in Q3. The number starts picking up next week.)

October buyback activity has been running about 40% below the year-to-date average, meaning corporate America sat out most of this sell-off. 

But, when they re-enter, their ammunition is going to be bigger than ever.

Buyback authorizations have hit a record $1.3 trillion so far in 2026, up from $1.2 trillion for all of 2025. 

And these aren’t just announcements.

S&P 500 gross buybacks grew 15% y/y in the first half, with growth accelerating to 20% in Q2.

Now, we should expect less buybacks because hyperscalars are spending free cashflow on GPUs instead of buybacks. But, that means own semis.

So, from what it seems, corporate America has $1.3 trillion and is waiting for the bouncer to open the door.

When the rope drops Monday, it isn't far-fetched to expect it's Black Friday at Best Buy.

(We may see the midterm hangover last till November 3 before things get back on track.)

October 26 is also the day markets bottomed in 2023. We’ll take the coincidence. Long-term rates also continue to drop.

The third reason is that the crowd has already gotten nervous.

The AAII bull/bear ratio has dropped to 0.81, well below its long-run average of 1.18. BofA’s bull/bear has fallen to 7.2 from 9.5 in Aug '26.

The CNN Fear and Greed is now at 21, in extreme fear territory. 

Investors are nervous about the markets and are staying away. Their comeback should fuel markets higher, which we see likely as earnings season progresses.

We’re leaning into momentum and quality now.

Momentum got hit hardest in the recent sell-off.

Goldman’s High Beta Momentum pair has erased all its gains for the year after a 43% drawdown from its June peak, when it was up roughly 70%.

Unconstrained Momentum is in its worst drawdown since the financial crisis, excluding Covid.

Momentum factor included names in semis and memory, which sold off mainly due to political risks around datacenter constructions.

This week, Amazon scrapped its plan to build data centers on George Washington University’s Northern Virginia campus after widespread opposition from residents.

(The NIMBY is unfortunate. It also shows the success of Russia / China propoganda and luddites of various types. Did you see how the bot spam traffic dropped considerably on X after Ukrain took out 3 Russian datacenters?)

Loudoun County is known as “Data Center Alley.” And, the backlash there means it is going to be a pressing issue elsewhere as well.

With midterms approaching, data centers have become a political issue, and investors sold the AI trade on fear that the backlash spreads.

But that’s exactly the kind of sell-off you want to buy.

Midterm fears peak right before the vote. Once the election clears, the uncertainty goes with it.

Historically, buying dips in the momentum factor has worked, and positioning is now much cleaner after an extreme drawdown.

Meanwhile, demand for compute keeps rising. Look at the post from Deedy Das (Partner, Menlo Ventures) below.

His point is that inference is printing money for anyone with GPUs.

Rent a B300 at $8 an hour, run an open-source model on it, and you can earn $10 to $20 an hour in token revenue.

DeepInfra can earn $38 an hour on DeepSeek during off-peak hours.

New inference companies are popping up every week, and the pie is growing so fast that even the upstarts are seeing revenue skyrocket.

Every one of them needs more compute.

One cancelled campus moves capacity to another county. It doesn’t make the demand disappear.

AI Value Capture is Shifting Away from LLMs

That said, we would be very cautious on Anthropic and OpenAI… it’s so incredibly easy to swap these models with a US-hosted version of DeepSeek. DeepSeek is at 90% of the cost and is ‘good enough and getting better’.

This is why Anthropic revenue growth is slowing…

I’ll have much more on this next week. Where you want to position in AI is around Harnesses and AI Agent Orchestration. They own the engineer desktop now, and the LLMs are becoming commodities or airlines.)

Quality Factor Is Cheap Too

Quality factor sold off too, though it has recovered from its lows over the past month. The market is broadening: the S&P 500 excluding AI stocks outperformed the S&P 500 by 6 percentage points in October.

Both these factors are looking good heading into the year-end.

You can see how these factor based strategies act in the Lumida Invest app, which tracks how each of our strategies is performing in real time. 

You can also follow any strategy in the app and see exactly what it’s buying and selling as it rebalances. Download the Lumida Inevst app here.

We will be adding Video Avatar agents there soon - they are already there if you have the testflight version.

Meta reported earnings on Wednesday, and the stock fell 7%.

The quarter itself was strong. Revenue rose 26% to $64.6 billion.

Headline operating income fell 37%, but that was because of a $9.9 billion legal settlement. Excluding it, operating income grew 11%.

So revenue didn’t break the stock. Capex, or more particularly where the funds for the capex come from, did.

Meta spent $34.9 billion on capex in the quarter. It generated roughly $33 billion of operating cash and spent all of it, plus a little more, leaving free cash flow at negative $2.1 billion. 

It also raised the bottom of its 2026 capex guidance, now $135 to $145 billion.

Meta noted that recent capital raises from Alphabet and SpaceX, along with expected raises from Anthropic and OpenAI, “have created an increasingly challenging fundraising environment.” 

And, Meta isn’t the only one saying this. 

Hock Tan (CEO, Broadcom) recently warned that capital could become the biggest constraint on AI. 

Larry Fink (CEO, BlackRock) called the competition for capital the fiercest of his career. 

And on Thursday, Jensen Huang said neocloud demand is “off the charts,” but capital is scarce.

These comments made the market question: Will hyperscalers have the funds to finance the earnings boom that was driving the semis, energy, and other market pockets? 

The sell-off spread quickly. Nvidia fell 4.4%, CoreWeave 10.1% and Lam Research 11.8%, while the semiconductor ETF (SOXX) dropped 6.8%. 

But, is Meta’s negative cash flow really an indication of a systemic problem?

We don’t think so.

Hyperscalers, like Meta, Google, Amazon, are reliable cash-generating businesses that can fund this buildout from cash flows because they generate great returns on their investments. 

Meta’s negative free cash flow print looks bad. But underneath, it doesn’t look as bad. 

Meta ended the quarter with $82.4 billion in cash, and its core business is accelerating.

On Instagram, time spent per user hit an all-time high and grew double digits. AI-based relevance improvements lifted Reels time spent by more than 4% in the quarter. 

The newer AI ad models drove a 9.6% increase in Meta’s app-ad conversions. Partnership ads nearly doubled to a $15 billion revenue run-rate.

Susan Li (CFO, Meta): “Our AI investments are already delivering substantial incremental advertising revenue.”

Every dollar of compute Meta puts into ranking and ads is still earning a return, and Q4 revenue guidance of $73 to $76 billion (~+28% YoY) says so.

Then there’s Muse, which is the reason Meta needs all that compute.

More than 10 million people in the U.S. and Canada set up a Muse personal agent in about six weeks. 

Only 11 other apps have hit 10 million U.S. downloads that fast. About 25% of users come back daily, which Zuckerberg said is more than twice where Threads started.

Muse is now ahead of ChatGPT on downloads and daily users at the same point after launch.

Citi estimates about 20% of Muse users already generate more than $1,000 a year in revenue across Meta’s apps. This shows the adoption can increase as users are more profitable, and able to fund expansion.

So how should you read this?

Meta has a core business compounding at 26%, a cash pile to fund the transition, and a consumer agent with real product-market fit. 

Having said that, we aren’t buying Meta here.

The fight over capital is real, and negative free cash flow will keep the stock volatile until Muse’s revenue starts to show up.

It’s better to own the names that Meta is spending their free cashflow on. That would include Nvidia, Taiwan Semiconductor, and others in the semi supply chain like Applied Materials and memory names.

Every dollar Meta spends on capex is a dollar of revenue for the companies selling it chips, memory and equipment.

That’s where we’d rather be. We like Founder CEOs like Zuck - but we think better to follow the money.

I discussed the debt concerns around neo-clouds, especially Coreweave, and the demand for compute in this FSD. Watch on our Youtube channel.

Meta shareholders carry the burden of the spending. Semis shareholders collect it. And after this week, semis are cheaper. 

SOXX is down 6.8% on a report that raised the floor on capex and said the core business still has room to absorb more compute.

Anthropic’s S-1 will probably be the least-revealing S-1 in a very long time. 

Everything that an investor might want to look at is already out. 

This shows Anthropic has got a clever CFO and/or bankers. 

They leak some of the financials to the media. Assess reaction. Prepare responses, adjust narrative. 

And, honestly, when the stakes are two trillion dollars, this is the right way to do it. 

You have a teaser to gauge reaction. They also have good surrogates like Gavin Baker and Altimeter out there. 

OpenAI should take notes.

The most recent headline on Anthropic's numbers is the loss.

Anthropic recorded an $80.7 billion net loss in the first half. While this number is big, and everything Michael Burry would want investors to look at, about $72 billion of it is mark-to-market losses from convertible notes it used to raise money.  

As Anthropic’s valuation rises, those notes are worth more, and the accounting loss grows with them. 

The more investors love the company, the bigger the loss gets.

Anthropic is reportedly eyeing the IPO valuation at about $2 trillion.

Here’s what $2 trillion looks like.

At the $2T price, Anthropic would be worth more than 35 SaaS companies combined, including Salesforce, SAP, Adobe, ServiceNow and Intuit. 

Together, they are valued at less than half of Anthropic’s expected valuation, and generated 4.6x Anthropic’s 1H26 revenue.

While we are on the comparison, let’s compare it to Meta.

After this week’s sell-off, Meta is worth roughly $1.7 trillion. 

Meta did $64.6 billion of revenue last quarter, growing 26%. Anthropic told investors its third-quarter revenue was $16.8 billion, roughly a quarter of Meta’s.

Meta has 3.6 billion people using its apps every day, a core business that’s still accelerating, and a consumer agent that hit 10 million users in six weeks. 

Anthropic has impressive but slowing growth, revenue concentrated in a few customers, and a compute supply that depends on its competitors.

The public market just sold Meta down 7% for spending too much. The private market wants to price Anthropic above Meta with likely bigger spending obligations and lesser outlets to fund those.

That gap tells you how stretched private markets have become.

If you’re an early holder, it’s probably a good idea to do secondary sales before you’re in a lock-up.

(I’ll explain next week where value capture is going in private markets.)

And if you want exposure to Anthropic’s growth in public markets, you already have it. That $16.8 billion quarter goes straight into compute, chips and memory.

Anthropic isn’t the only lab raising at a trillion-dollar price.

OpenAI is nearing a $30 billion funding round at a $1.25 trillion valuation.

UAE investors and Coatue are leading the round, which is being positioned as a bridge to a potential IPO in 2027. 

This is the capital race Zuck was talking about.

Every lab is going back to the same pool of money, and the checks keep getting bigger.

The difference is that OpenAI is raising from a relatively weaker position than it had a year ago.

It had a massive lead and lost it to Anthropic in the market that pays the most: enterprise and coding. 

Claude Code became the default tool for developers while OpenAI was still focused on consumers.

Now China is coming from below. 

DeepSeek is 90% cheaper than the frontier models, and the open-weight models are good enough for most tasks.

OpenAI is getting squeezed from both sides: a better competitor at the top and cheaper ones at the bottom.

It’s The Game Of Distribution Now.

When models start to look alike, the model stops being the product.

What wins is distribution.

Meta just showed what that looks like.

Muse got 10 million people to set up a personal agent in six weeks. It didn’t launch on the best model in the world. 

Muse won because Meta already had 3.6 billion people using its apps every day. It never had to go find its customers.

OpenAI seems to have realized this, and has signed a distribution deal with Google. 

OpenAI will bring its frontier models, including GPT-6 Astra and Codex, to Google Cloud in the coming weeks. 

Enterprise customers will be able to buy them through Google Cloud Marketplace, using cloud spending commitments they’ve already made.

That last part is the key. A CIO doesn’t need a new budget line to buy OpenAI. The money is already committed to Google.

The deal follows OpenAI’s expansion to AWS after its exclusive Microsoft arrangement ended. 

OpenAI is now on all three major clouds.

That gets OpenAI reach. The question is who owns the customer.

Think about how Microsoft’s Copilot works. It rations queries across models, sending each task to the model that makes the most sense for it.

If Google does the same, the harder queries go to OpenAI and the easier ones go to Google’s own Argon. Most queries are easy.

Over time, people get used to using AI inside Google’s product suite. They don’t think about which model answered. They think about Google.

That habit is the valuable asset.

The customer relationship sits with Google. OpenAI becomes the engine under the hood, paid for the hard queries.

The model isn’t the product. The experience is, and whoever controls where the customer already spends their time controls the experience.

Meta, Google and Microsoft own that distribution. The labs are renting it.

That’s fine as a business. But it’s hard to justify a trillion-dollar valuation as the supplier inside someone else’s product, especially when cheaper suppliers are lining up behind you.

You can now view Lumida Curations at the Lumida Invest App. Curated insights delivered on time. Download the app here.

Elon Musk argues that digital superintelligence would lift global output by only about 20% to 30%, and that the real transformation comes from robots and space, which could grow the economy to a million times its current size.

View Curation

Ken Griffin argues that GPUs, like airline seats, cost almost nothing to run at the margin, so any overcapacity could send compute prices into free fall, just as unlit fiber crushed telecom pricing after 2000.

View Curation

Dylan Field, co-founder and CEO of Figma, argues that AI is making digital creation easier for everyone, but that only people willing to focus and take bold risks raise the ceiling on what great work looks like.

View Curation

Not Subscribed Yet? Don’t miss out on future insights—subscribe to the newsletter now!

For real-time updates, follow us on: 

X | Telegram | YouTube | TikTok | News | Ram’s X | Lumida Health | Lumida Tax | Lumida Memes

As Featured In

Disclaimer: Lumida Wealth Management LLC (“Lumida”) is located in New York, NY, and is an SEC registered investment adviser. SEC registration does not constitute an endorsement of the firm by the Commission nor does it indicate that the adviser has attained a particular level of skill or ability. Lumida only transacts business in those states in which it is registered, or qualifies for an exemption or exclusion from registration requirements. Any direct communication by Lumida with a prospective client will be conducted by a representative that is either registered or qualifies for an exemption or exclusion from registration in the state where the prospective client resides.

The information in this material has been obtained from sources believed to be reliable. While all reasonable care has been taken to ensure that the facts stated in this material are accurate and that the forecasts, opinions and expectations contained herein are fair and reasonable, Lumida, Inc. and Lumida Wealth Management LLC (collectively Lumida) make no representations or warranties whatsoever the completeness or accuracy of the material provided, except with respect to any disclosures relative to Lumida. Accordingly, no reliance should be placed on the accuracy, fairness or completeness of the information contained in this material. Any data discrepancies in this material could be the result of different calculations and/or adjustments. Lumida accepts no liability whatsoever for any loss arising from any use of this material or its contents, and neither Lumida nor any of its respective directors, officers or employees, shall be in any way responsible for the contents hereof, apart from the liabilities and responsibilities that may be imposed on them by the relevant regulatory authority in the jurisdiction in question, or the regulatory regime thereunder. Opinions, forecasts or projections contained in this material represent Lumida’s current opinions or judgment as of the day of the material only and are therefore subject to change without notice. Periodic updates may be provided on companies/industries based on company-specific developments or announcements, market conditions or any other publicly available information. There can be no assurance that future results or events will be consistent with any such opinions, forecasts or projections, which represent only one possible outcome. Furthermore, such opinions, forecasts or projections are subject to certain risks, uncertainties and assumptions that have not been verified, and future actual results or events could differ materially. The value of, or income from, any investments referred to in this material may fluctuate and/or be affected by changes in exchange rates. All pricing is indicative as of the close of market for the securities discussed, unless otherwise stated. Past performance is not indicative of future results. Accordingly, investors may receive back less than originally invested. This material is not intended as an offer or solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to particular clients. The recipients of this material must make their own independent decisions regarding any securities or financial instruments mentioned herein and should seek advice from such independent financial, legal, tax or other adviser as they deem necessary. Lumida may trade as a principal on the basis of its views and research, and it may also engage in transactions for its own account or for its clients’ accounts in a manner inconsistent with the views taken in this material, and Lumida is under no obligation to ensure that such other communication is brought to the attention of any recipient of this material. Others within Lumida may take views that are inconsistent with those taken in this material. Employees of Lumida not involved in the preparation of this material may have investments in the financial instruments or securities (or derivatives of such financial instruments or securities) mentioned in this material and may trade them in ways different from those discussed in this material. This material is not an advertisement for or marketing of any issuer, its products or services, or its securities in any jurisdiction.