In partnership with

The voice-to-text app that built its own speech model

Most dictation apps are a wrapper around someone else's speech model. Wispr Flow trained its own.

It's called Canto, and it was built for the way people actually talk to AI: thinking out loud, changing direction mid-sentence, half-whispering in an open office.

Wispr tested it on 10 hours of real dictations from more than 2,300 speakers. Canto had the lowest word error rate of every model tested, including models from Google, OpenAI, AssemblyAI and Deepgram. That means your prompts to ChatGPT, Claude and Cursor come through the way you said them.

Talk naturally in any app and Wispr Flow types clean, formatted text, 4x faster than typing.

Good morning partner,

A few things to cover this morning:

** PREMIUM MEMBER REMINDER (If you haven’t yet) ‼

You can now access the Telegram Community to receive our live trade desk alerts in real time (no more waiting on emails to hit), and a dedicated Q&A space with access to Gabriel for anything you’d like to discuss

Yesterday’s Session

We have now officially entered Phase III out of the three phases I’ve been describing all month.

The brokers who bought the forced liquidated assets from the initial AI selloff now are seemingly dumping the positions they didn’t want on their books in the first place.

Why am I so quick to assume that?

We just had a market-wide de-risking where all but one sector was down on the day, all ahead of the NVIDIA earnings announcement tonight.

Can Jensen make a surprise comeback? Will Trump address the nation and markets with some sudden relief on the Middle East or a wild card?

Both are possible and probable, but then here’s the other counterargument…

Retail trading volume hit a record on this dip-buying, likely leaving many in the same position that kickstarted this three-phase market structure.

Another such selloff could trigger another round of forced liquidations and margin calls, and that’s just on the financial side.

For individuals who do not invest, they are all being distracted by this “global power outage” narrative, forcing them to tighten their cash flows into buying survival kits and other essentials.

It just seems like the Anthropic IPO + NVIDIA earnings mix this week is the designed riddance of the individual investor through exit liquidity.

Whatever happens though, I think it is noise, but it’s useful noise.

We see names in AI fail to rebound from the lows and reclaim their old “bear market” levels where a 20% sell from the highs was seen.

At the same time, many other non-AI assets have gained 10-20% off their lows, sending them into official bullish corrections and markets.

A couple of defensive names in our portfolio have staged such a recovery, giving me confidence that the only places money wants to run into are these type of companies.

Not to say they are off the hook in a broad de-risking, but their relative price action does strengthen my case for them.

Let’s take a look at Yesterday’s leaders and laggards:

Defensives were the only sector that managed to escape the carnage.

Confirmation into those relative swings off the lows is being made at the sector rotation level.

While the market is quick to dump everything related to AI, and even throw in some “growth” shock de-risking sentiment…

A defensives and somewhat real estate leadership works wonders for you to see who is left standing in the panic.

I invite you to see how well your holdings are doing both off the highs (corrections? bear markets?) and off the lows as well.

Especially during broad de-risking like this week.

We gave Premium members a particular opportunity yesterday, in a company whose market treatment is significantly detached from reality to create a relatively easy risk/reward opportunity to squeeze profits from.

I believe once the realization sets in that this stock should be treated differently than it is today, our payoff will slowly come in.

It’s Probably Time

This weekend, I will be hosting a free webinar covering three very important things…

  • Why the market mechanics have changed over the past decade

  • Where money is still being made now (with proof from our own booked operations)

  • What the next decade has in store for equity investors

When looking at each period between 1910-1930, 1970-1980, 2000-2010…

The famous “lost decades” in the market,

Plenty of indicators had all the similarities we can see in today’s economy and S&P 500.

These were also the dates when the value investing legends decided to step away from the business:

  • Warren Buffett: Closed his partnership in 1969.

  • Julian Robertson: Closed Tiger Management in 2000.

  • Peter Lynch: Retired in 1990s.

  • Joel Greenblatt: Returned all outside capital in 1995.

They didn’t suddenly lose their skills or grip on the market…

Long-only investing simply doesn’t work during markets like these unless you pivot into the right areas,

Areas where institutions and fund managers cannot play around in.

That’s where I would put my money to work in during the next decade, and the data proves this is the most advantageous pivot you could make in your career.

I will break all of this down this weekend, for free.

News

Movers & ES Levels

  • Xpeng 📈 Rose over 3% after posting a better quarter than expected, continuing to boost the Chinese thesis I have now on the table. Personally, I’ve been seeing more BYDs and Xpengs than Teslas.

  • Molina Healthcare 📈 Popped 11% after states announced they will extend and/or delay work requirements for Medicare participation, boosting the amount of premium revenues this defensive sector gets.

  • Robinhood Markets 📉 Lost 7% after prediction markets competitor Kalshi announced it will step into the stock trading and investing business. I don’t think customer migration will be that great given the brand penetration Robinhood has achieved. Here’s my price target for it.

  • Intel 📉 Slipped 7% after an investor lawsuit claiming the company failed to communicate the extent of the PC market and sales slowdown, I was telling several premium members to get out of this stock weeks before.

Now let’s get into some ES levels for today.

We have lost $7,525 and even $7,500 for a massive selling event into the $7,400 mark.

If you are a buyer on the NVIDIA earnings sentiment, I would not want to do anything here unless the index can reclaim and hold $7,430 or so.

Past that, there’s a lot of ground to cover if bulls want to see this tape repaired.

First of all, there’s the $7,450 mark where a lot of “trapped” aggressive sellers sit from yesterday.

If there’s anyone left to buy during this forced liquidation, that would be the price to show their hand.

Then we meet a great amount of resistance around $7,500 as a key level to reclaim here.

Somehow getting past that could get us back into the $7,550-$7,575 range and look to close the week there to make higher prices possible.

If you’re trading today, selling $7,430 is your best option and likewise buying a breakout or breakout and retest of $7,430 is your best option.

Greatly divided market for today so be careful, tread lightly.

Portfolio

Even with a market selloff and AI stocks digging their heels in a bear market…

The portfolio is only ~2% off the all-time highs.

Which goes to reiterate the fact that it doesn’t matter how much money you make, but how you make it.

Some traders are scratching on 30-40% gains, some are fighting to be at breakeven.

The common trait between them is they can see 5-8% daily swings in this market and lose sleep, and having very little cash to take action when necessary.

We see a 2% day as highly volatile, and still have roughly 60% in cash left to deploy on the best opportunities once the lows begin to show themselves.

It’s just how we do it over at Premium.

Here’s the positioning update for today’s premium members:

logo

Subscribe to our premium content to read the rest.

Become a paying subscriber to get access to this post and other subscriber-only content.

Upgrade