#90 Apple’s Intelligence, Smartphone Proxy Wars

+ Roaring Kitty Litter, The Snowflake Snowball, Karmic Debts, Elementary My Dear Botson

The Week that Was:

A look back at the tech world of the past week.

Apple’s Intelligence:

So we finally got Apple’s 1 hour and 45 minutes long Worldwide Developer’s Conference (WWDC) showcase and shortly after the opinion pieces and breakdowns came flooding in. We had thrilling headlines such as “Math notes was the most exciting WWDC announcement.” Wait…what?

Yeah, the opinions were decidedly mixed on the face-value contents of this WWDC, with my own succinct opinion being ‘boring but practical.’ The math notes application referred to here was an integrated calculator solution that recognizes and accounts for handwriting and drawing across Apple’s suite of apps, with API support for third-party developers' integration.

This was noteworthy because of the degree of precise, seamless control it demonstrated over various intertwined machine-learning techniques. Not only did early generative AI offerings do a notoriously poor job of basic math, but current generative AI still struggles tremendously with handwritten text.?

With the obvious caveat that this was a carefully rehearsed, ideal circumstance preview, math notes performed both flawlessly and intuitively. It was the tip of the spear for a wider suite of AI-powered productivity tools integrated across Apple’s entire ecosystem. I’m not going to talk about that in further detail because you already know all of it. Not just because Bloomberg leaked the entire showcase 4 days in advance, but because literally none of this stuff is new.

That’s what made it so boring. And therein lies the big reason why I disagree with those that consider this showcase a waste of time. Slightly more than half of the showcase was dedicated to Apple Intelligence, with Tim Cook kicking things off with a certified “say the line Bart!” moment where he said “Artificial Intelligence” for the first time since the generative AI mania kicked off.

Almost all of the Apple Intelligence stuff shown was stuff that my mother and sister could and likely would use. They are smart, capable people. They are not particularly into tech. This is what has prevented AI from breaching the ‘normie barrier.’

Despite all the hype and investments, AI is still niche in the grand scheme of things. Apple’s play is to mainstream AI. Once it’s finally ready to do so that is…Like practically everything from this showcase, Apple Intelligence is ‘coming soonTM.’ And if it sticks the landing, oh boy. LINK

Smartphone Proxy Wars:

To be blunt, we won’t know shit about how Apple Intelligence works in practice until July at the earliest. This is when the earliest developer previews might begin. But we can already intuit quite a bit. The Verge’s David Pierce did just that, with a beautiful encapsulation of Apple’s priorities.

“Really interesting how much appeal for him (Craig Federighi) here is UI: an integrated chatbot is just easier and faster to do stuff. I think that’s true, but it’s definitely not a “this will transform everything about life forever!” case. It’s just a better way to use your phone.”

It’s not sexy, it’s not gonna blow minds or investor budgets, but it is deeply, eminently practical. Challengers to the smartphone continue to break themselves upon the smartphone. Apple’s core business is to sell smartphones. It desperately wants Intel MacBook stragglers to upgrade as well.

“just a better way to use your phone” is far more tantalizing and tangible a prospect for the average smartphone user. It also comes at an interesting time amidst mounting pressure by various interest groups to try and restrict smartphone use amongst children.

Fears over child smartphone use are hardly, new. In fact, I’ve long been writing about resultant trends like parents buying smartwatches for their kids instead of phones.

Humane and Rabbit sought to capitalize on such trends but failed to understand the core drivers behind them. Namely, that people just want a better way to use their phones, not a way to replace their phones.

In that sense, it’s absolutely not a coincidence that controlling one’s iPhone from a Mac is one of the marquee new features of iOS 18 and MacOS 15. It’s not about using your iPhone about your Mac. This feature is about laying the groundwork for full usage of your devices, without needing to directly interact with them.

It’s simply not feasible to cram enough power into glasses or watches to let them replace your smartphone. And why bother? I am confident, the likes of Samsung, Google, and Huawei are also working along these same lines of reasoning. With their answers to Apple’s play coming soonTM. LINK

The Business End:

Coverage of big business moves and events in the tech world.

Roaring Kitty Litter:

You might be wondering why prominent financial institutions and publications keep floating headlines lately that worry about dollar dominance.

Analysts writing such pieces worry that the present-day American mindset is one of arrogance and laziness. It is their fear that a deep-rooted ‘culture of indolence’ will cause the US’s technological and financial dominance to slip from its grasp.

To back up such claims, they point to cases like the ‘roaring kitty.’ This is a random dude who a notorious subreddit (/rwallstreetbets) formed a cult of personality around. ?And I would argue that he’s a symptom, not a cause of the massive market volatility his ‘followers’ antics caused. Rather than inherent laziness or arrogance, I think the real problem with the dollar is how the unicorn mindset has long outstayed its welcome.

In business parlance, a unicorn is a young company that manages to attain a massive valuation in record time. It is the myth of American exceptionalism wrapped up in the thrilling narrative of an underdog beating ‘the man’ and making it big through sheer ingenuity and hard work.

To ground such myth-making, there tends to be one central figure around whom a cult of personality forms. And the major fall-from-grace stories we’ve seen, such as Sam Bankman Fried’s Cryptocurrency collapse, are the natural result of these ever-loftier dreams inevitably colliding with reality. And it’s no coincidence that all these major seismic shifts keep happening in the tech sector.

Silicon Valley is the breeding ground of unicorns and the jewel in the US’s trillion-dollar crown. It is also the land where America’s dreamers need never wake. Roaring Kitty’s ‘failed’ stock rally is yet another wakeup call.

Reality may be more boring than the American dream of unicorns. But it is also a lot more practical. In reality, companies need to offer meaningful products and services as part of sustainable and rational business models.

Thus, I predict the present trend of prioritizing real product value and business model viability to continue, and for the dreamers to wake up sooner rather than later. LINK

The Snowflake Snowballs:

In cybersecurity, one of the most important elements to always keep in mind are so-called ‘attack vectors.’ This simply means ways and routes through which someone who means you harm can bypass your defenses to steal or break your stuff.

Fear of attack vectors is why the city of Amsterdam is removing Chinese-made cameras, it is also why security professionals in particular are weary of deep integration of AI agents into operating systems.

Attack vectors abound these days, so small wonder we’ve seen some more high-profile security incidents being disclosed this past week. In the firing line this time are the New York Times, streaming platform Niconico, and cloud storage firm Snowflake.

Snowflake was actually a part of my cybersecurity items last week,? but since it wasn’t clear enough yet to what extent the company itself had been compromised at the time of writing, I opted to let that particular story marinate for a bit more. And here we are. A big ol’ breach snowball rolling down the hill.

Snowball denies being directly breached itself, yet the same infrastructure it offers as a core value proposition was nonetheless used by threat actors to breach Snowflake customers and their customers in turn. Hence the snowball analogy. Predictably, using single-factor authentication wasn’t a good idea for sensitive company data, who would have imagined that?!

Whilst Snowflake is now urging all customers to use multifactor authentication, one can’t help but wonder how sloppy security practices were even permitted to begin with. Simple, because sloppy security is more convenient.

People who dislike cybersecurity stories rightly complain that “It’s always the same thing. Criminals get more creative, so security gets upgraded. Some get hacked, and others promise to be more careful, so they don’t get hacked. Same old song and dance over and over!”

And yeah, it is, because user experience is the real weak link in all cybersecurity. We are now seeing these ‘mega breaches’ as a natural result of security and data retention consolidation. In the past, a breach might have been a single snowflake falling upon the data landscape. These days, a single Snowflake falling can snowball in moments to trigger an avalanche. LINK

Rules of Engagement:

Ethics and legal matters regarding tech engagement.

Karmic Debts:

What is Karmic Debt? This is part of a personal business philosophy of mine which I periodically repeat within these newsletters. Simply put, stakeholder sentiments such as goodwill and loyalty are currencies, the same as money. Karmic Currency (KC) is the umbrella term I use for them.

Like fiat currencies, KC can also be earned, spent, gambled, and squandered. However, unlike fiat currencies, KC can’t be topped up by flirting with venture capitalists or other investors. The more KC you have, the wider your margin of error as a company. The more KC you have, the more supportive and cooperative stakeholders will be towards you. And so on and so forth.

Basically, it’s not immediately visible on spreadsheets, but having high KC reserves helps a company out in countless ways. So what happens when you blaze through all of it and end up in debt? Easy, no one wants to give you the benefit of the doubt anymore. No one will say ‘ok’ when you say, ‘trust me.’

Why is it that, despite considering it boring, people are so much more receptive towards Apple Intelligence vs Microsoft Recall? They are exactly the same thing. And yet coverage of Apple’s AI is cautiously optimistic, whereas people furiously recoiled from Recall. In fact, Arstechnica literally wrote a headline stating, ‘Windows Recall demands an extraordinary level of trust that Microsoft hasn’t earned.’

This is why many other tech companies like Adobe have also been on the defensive lately. They’re all in karmic debt. Up to a point, karmic debt doesn’t have to be crippling. However, if you need to compete with rivals in a neck-and-neck race, suddenly karmic debt will hobble you.

A case in point would be if Apple Intelligence (eventually) arrives, does what it promises, and gains exclusive, privileged access to intimate data. The kinds of data that stakeholders are presently prepared to fight others to the death for.

So that’s my advice when looking at who might win any given tech arms race, look at their karmic debt before considering their market cap. LINK

Elementary My Dear Botson:

So apparently calling someone a ‘bot’ online is considered an actual insult now? Could have fooled me, I thought it was still just mean-spirited GPTeasing at most. Digging a little deeper though, it’s not just calling other people bots that appears to be a mark against their intellectual rigor and narrative wit. Supposedly writers who use AI writing aids also accept lower pay for their services.

That’s kind of dark considering how underpaid African workers are still widely used to moderate social media platforms and refine datasets. It also confuses me because access to AI is seen as a mark against someone’s education and credentials. This is despite the fact that only the more affluent and well-connected (literally in terms of internet stability) among us can even use such tools right now.

And it’s not just AI access that is unequal, all tools are, even free ones like open-access publishing. Hence why researchers in lower-income countries enjoy fewer benefits. You can’t use something that you can’t access.

Now, there are obviously geopolitical motivations at play here as well, but China’s renewed efforts to reach out to African students with scholarships are still commendable. Despite the increasingly repressive nature of Chinese life, its academic institutions continue to be some of the most productive in the world, with an admirable combination of quality and quantity given their relative size.

Little wonder then that it's doubling down on this core strategic asset. Kind of like how the US has to let go of its Unicorn dreams for the dollar to remain on top, China’s academic dominance will live or die based on whether or not the CCP can learn to let go of its more egregious control freak tendencies. It’s not the only one with this problem, mind you, but it does suffer most from intellectually self-harming policies.

Currently, too many African students return home with mixed results from their studies abroad because of that. The ‘yields’ so to speak, simply aren’t good enough. Increasing these yields will benefit Chinese foreign policy, but also force it to ‘let go’ more than it might like. LINK

A Nice Cup of Serendipity:

Cool bits and bobs from around the web.

Space Wood LINK

The Hero LINK

Blood Sugar LINK

Lessons From Doge LINK

Twinning LINK

Ikea x Roblox LINK

Old Ecosystems LINK

Computex roundup LINK

ASML’s New Toys LINK

Shroom Hunters LINK

Wholesome Recipe: LINK

The Deep End:

A weekly batch of long-form content recommendations.

Apple’s Foundation Models:

A peak at what might be coming soonTM? to an Apple device near you. LINK

Summer Game Fest:

An overview of one of the many gaming showcases from this past week. LINK

Baldur’s 3 Cinematics:

A Deep dive into the cinematics of this award-winning RPG. LINK

One More Thing…

In case you were wondering why I have ‘coming soonTM’ repeating across this newsletter, it’s my little way of poking fun at the way Apple’s presentation was literally 80% ‘coming soon.’ It was so bad that a colleague approached me at work to ask “Hey did I install the wrong iOS 18 developer preview? Where is everything?”

At the latest, we are looking at being well into 2025 before all of the announced stuff from WWDC 2024 is feature-complete on consumer devices. At that point, Apple Intelligence - the slowest to roll out of the whole lot - would have been a better fit in ?WWDC 2025.

I am adamant though, that if Apple actually does manage to deliver on even half of its WWDC 2024 promises, most, if not all its AI rivals will be dead in the water. Like I said they would be in 2022 and 2023 if they ignored on-device compute and privacy!