beyond the alphabet
Beyond the Alphabet is a weekly column focused on the world of technology in and around Mountain View.
I/O 2024 is just around the corner, and Google just made an announcement that’s probably bigger than what’s coming next month. Rick Osterloh will no longer just run Devices and Services, he will lead the new “Platforms and Devices” team. The reorganization will see Osterloh continue to oversee his Pixel team, with responsibilities for Android, Chrome, ChromeOS, Photos, and more.
Without listing all the different divisions, it might just look like a surface-level change within Alphabet. However, there’s another big change as Hiroshi Lockheimer begins “working on several new projects across Alphabet.” While it’s nice to see Hiroshi remain with the company, it’s a little strange not to see him at the helm of his Android and his ChromeOS.
This isn’t something we haven’t heard before
In a statement to The Verge, Osterloh explained that AI was at the center of this dramatic restructuring. Considering how determined Google is to push AI and Gemini into everything it can, this integration makes a little more sense. But I can’t help but feel like I’ve heard this before.
When Google announced Pixel 6 with Tensor chips, the assumption was that the hardware and software teams would work together. Essentially, the Pixel lineup is given the same treatment as the iPhone, with the SoC being tweaked and tuned to match the hardware it has on board.
Both sides have to debate whether Google has followed through on the proposition it laid out just a few years ago. But then OpenAI and ChatGPT lit the fuse, making AI available to consumers. Google failed to launch Bard and later rebranded it to Gemini. Google, on the other hand, continues to stumble a bit, as evidenced by claiming that the Pixel 8 won’t feature Gemini Nano, only to retract that claim.
AI gaming is still in its relative infancy, so all hope isn’t lost for Google. But Osterloh needs to grab its horns, so to speak, and actually solve the problems that still plague its users. That means turning Google into the leader it was meant to be and stopping Samsung from stealing all the credit.
Perhaps this is the fundamental change needed to avoid problems like the Pixel 8’s Gemini Nano. Don’t make users of the most expensive Pixel smartphones feel like second-class citizens in exchange for a promise of seven years of updates. Instead of acting as if you are just a temporary resident, act as if you own the place.
Is this organizational reorganization really due to AI? Probably not, but that’s Google’s story and they’re sticking to it.
This isn’t Google’s first foray into consumer-facing AI products and features. Pixel smartphones have long ranked among the best Android smartphones, even though their camera hardware has been laughably poor. This was all made possible by the machine learning models that were (and are) used in Pixel phones, allowing you to capture clear photos of the starry sky.
The difference now is that Google is trying to implement AI more systematically without being seen as a laggard. Rumors and speculation suggest that Apple is planning a big AI-focused announcement at an upcoming developer conference. But Google has an opportunity to stay ahead of the curve simply because I/O happens first.
If there’s one thing Google and Apple have in common, it’s that both Siri and Assistant are a little lost. You can’t reliably have your assistant set reminders. Gemini may seem great, but it lacks many of the basic features we expect. Sure, it’s great to have a Gemini help you plan and book your trip, but take the plunge and ask them to turn off the lights. Siri, on the other hand, is laughable at best, and I remapped the action button on her iPhone 15 Pro Max to use her Perplexity instead of summoning Siri.
Google Assistant has failed again. I was asked to remind myself to respond to the email within an hour. Do a Google search instead. Sigh. pic.twitter.com/Jz8qfxOqQKApril 18, 2024
One of my recurring thoughts is that Osterloh is like running a company (Pixel) within a company (Google) within a company (Alphabet). This idea became even more clear to me after reading the statement “Google research team focuses on computational photography.” On-device intelligence will also move to a new organization to provide deep AI expertise across platforms and devices. ”
In doing so, Osterloh brings virtually all of the phone maker’s functional institutions under his direct control. Basically, I think of this as Osterloh founding a “Pixel Phone” company without going down the startup path like Carl Pei did with his Nothing. Not to mention, there’s also the added benefit of having a team of people he’s already familiar with, combined with Google/Alphabet’s financial backing.
Something else may be at play
We knew that Google was working on implementing AI into the devices and tools we use every day. It was really only a matter of time, but there’s one more thing about today’s announcement that left me scratching my head. Qualcomm CEO Cristiano Amon issued the following statement to The Verge: “We look forward to working with Rick to deliver a cutting-edge Android experience powered by Snapdragon, not just in mobile, but across auto, XR and compute.”
Is the partnership between Google and Samsung over? Will Qualcomm make Tensor chips in the future? Is Google reaching for the “white flag” that marks the end of the Tensor program as we know it? Are you over-analyzing a single sentence? Time will tell, except for the last question.
If this is the end the current Tensor iteration has been something I’ve struggled to write about for a while. Last year, there were reports that Google was “postponing” the launch of its “first custom Pixel chip” until some point in 2025. Initially, this “custom Pixel chip” was planned to be a Tensor G3. It’s built by TSMC rather than Samsung, and aims to offer a true flagship experience that rivals the best of Qualcomm and perhaps Apple.
Are we witnessing the “beginning of the end” for Tensors? Or maybe Tensors will be the next to be overhauled?
It wouldn’t be completely foolhardy for Google to change direction and work with Qualcomm. That happened with the Pixel Watch 2, which ditched the improved Samsung Exynos SoC in favor of Qualcomm’s Snapdragon W5.
Perhaps Google is looking to strengthen its relationship with Qualcomm rather than working with Samsung on design. His upcoming Snapdragon X Elite chip appears capable of matching the Apple Silicon that Windows and ChromeOS devices have sorely needed for years. This means that Arm devices can deliver the same experience as non-mobile processors, while increasing performance combined with much-needed efficiency gains.
Perhaps these are all just long examples of how people try to think “outside the box.” We can’t help but think Google has even bigger plans for the future. If I’m correct, Google’s biggest announcement in years is just the first drop in the bucket.
