The unspoken obnoxiousness of Google’s Gemini improvements

​ Depending on your perspective, two very different stories are a-brewin’ right now here in the land o’ Googley matters.

In one corner, Gemini, the AI-powered chatbot Google galumphed into the world this year, is getting better! I’ve lost count of the number of assorted little improvements that’ve shown up for the tool as of late, as it relates to Android and its increasingly apparent role as the platform’s next-generation virtual assistant.

At the same time, that progression emphasizes an unpleasant and often unspoken truth around Google’s rush to get its jolly green generative AI giant everywhere imaginable: This thing was rolled out way too soon and long before it was ready. It’s being forced down our throats for the sake of Google’s business benefit and at the expense of our user experience. And it’s being awkwardly and hurriedly shaped into a role it wasn’t designed to serve in a rushed-out, piecemeal manner instead of in the thoughtful, meticulous way that would have made it much more palatable for those of us who rely on Google’s services.

That first fork is the story Google wants to tell. The latter one, though, is the reality most of us in the real world are experiencing.

And I hate to be the bearer of bad news, but I’ve got a sneaking feeling it’s only gonna get worse from here.

[No sugar-coating, no nonsense. Get level-headed perspective on the news that matters with my free Android Intelligence newsletter. Three things to know and try every Friday!]

Google, Assistant, and the rocky road to Gemini

Let’s back up for a second to set the scene around this saga — ’cause goodness gracious, is it a strange yet simultaneously very characteristic-seeming tale, for anyone who’s been watching Google long.

Back in 2016, y’see, Google launched one of its biggest company-wide initiatives ever — a saucy little somethin’ called Google Assistant.

It’s hard now to even convey just how big of a deal Assistant’s arrival was at that point and then continued to be over the years that followed, up until extremely recently. Much like a certain plus-symbol-involving service we shall no longer name (pour one out…), Google set out to make Assistant appear everywhere.

From early on, the service connected to countless other Google services — from its integration into every imaginable corner of Android and ChromeOS to its home at the core of all Google-associated smart displays, TV systems, and speakers. It came to Android Auto, even, and showed up as the branding behind all sorts of smart Pixel calling features.

Heck, Google’s presence at tech conventions turned into a literal Assistant playground for years, with endless plastering of Assistant branding everywhere you looked and — well, stuff like this:

Assistant became the common thread across all of Google’s high-profile products, and even that was still just the start of the company’s grand Assistant ambitions. A few short years ago, Google was laying the groundwork for Assistant to evolve into its own fully featured platform.

As a certain greasy-beaked tech philosopher put it at the time:

Assistant is the Google platform of the future. Whether we’re talking about Smart Displays, the Home Hub, or Android devices, the operating system is but a pawn in Assistant’s larger-scale and higher-stakes game.

It seems safe to say Google devoted more time, energy, and likely also money toward building up Assistant as a tool and a brand than any other effort in recent memory. And, not surprisingly, it worked! Those of us who spend our days within the Google ecosystem learned to depend on Assistant for all sorts of tasks, ranging from quick actions across Android to cross-platform memory storage and on-demand smart device control.

And then — well, you know what happened, right? ChatGPT showed up. The tech industry freaked out about its future. And everything related to Assistant went to hell in a handbag.

The Assistant-to-Gemini (d)evolution

The signs of Assistant flailing first started showing up mid-last-year.

I’d gotten message after message from readers of my Android Intelligence newsletter and members of my Intelligence Insider Community asking the same basic question: What in the world is going on with Google Assistant?

People were noticing that Assistant was acting oddly and becoming more and more erratic. Commands that once worked fine were suddenly leading to strange results. Sometimes, Assistant just wouldn’t answer at all — or would return random errors anytime you tried to summon it.

I was among the lucky who didn’t run into any such issues for a while, but that’s absolutely changed in more recent months. The many Assistant-associated gizmos scattered around my home — speakers, screens, and other “Hey Google”-responding gadgets in practically every room — have become exercises in frustration. Error after error, failed command after failed command.

Plain and simple, what was once a reliable tech tool has turned into a steaming hot mess of disappointment. Even my kids, once obsessed with what they saw as the all-knowing and omnipotent Google genie, have now taken to berating and insulting the brand with inspired barbs like “Hey Google, why are you so stupid?” and — more direct yet — “Hey Google, you suck.”

While Google has yet to officially give us any guidance about its plans for the future of Assistant and how Gemini might fit into that, it’s become increasingly apparent over time that the company’s moving away from Assistant and devoting its time and resources toward building up Gemini as its replacement.

And that brings us to the other side of our current unchosen reality: When Gemini first showed up as an “experimental” Assistant alternative on Android earlier this year, it was an underwhelming glimpse at a future precisely none of us asked for — and that’s putting it mildly.

As I wrote back in February:

The real problem with Gemini as the Android assistant is that Google’s forgotten why a phone assistant actually matters — and what we, as actual users in the real world, need from such a service.

Using Gemini in place of Google Assistant feels like having a square peg awkwardly forced into a round hole. It feels more like an awkward adaptation of an AI chatbot than a phone assistant — something that’s half-baked at present and not at all intended or appropriate for this context.

And the more time you spend using Gemini, the more apparent that disconnect becomes.

To Google’s credit, in the time since then, Gemini as an Android assistant has only continued to get better. It’s clear that Google’s working to fill in the gaps and bring the service up to speed with Google Assistant as quickly as possible.

Heck, this week alone, we’ve seen signs of Google cookin’ up a Gemini automation system to take the place of Assistant routines, an improvement that’d allow Gemini to interact with streaming apps and control audio playback on your phone (something that, yes, has thus far not been possible), and a series of under-the-hood enhancements that’d make Gemini faster and more “natural” to interact with on Android.

Hey, that’s great! All of it. (Remember our diverging dialogues from a minute ago?)

But more than anything, it highlights just how badly Google has bungled this transition — and how much trust it’s rightfully gonna cost it in the eyes of its most committed users.

The Gemini on Android reality

The truth is that using Gemini as an assistant on Android, to borrow a phrase from my wise progeny, still sucks. And it’s not because its generative AI parlor tricks aren’t up to snuff. That stuff — all the new tricks Gemini brings to the table — honestly doesn’t matter all that much for most of us in this context.

As I mused a couple months back:

When it comes to an on-demand mobile device assistant, we don’t need the ability to have mediocre text or creepy images created for us from anywhere across Android. We need a fast, consistent, reliable system for interacting with our phone and other connected devices, getting things accomplished with our core productivity services, and getting short bursts of basic info spoken aloud to us in response to simple questions.

What Google’s scrambling to do now is catch Gemini up with those foundational basics — the table stakes, in other words, and the bare minimum of what makes for an effective and reliable phone assistant. And while it may be making impressive strides toward that goal, in the meantime, Gemini continues to fail at the core tasks we actually need from a service of this nature while also continuing to be pushed as a default for more and more people who never asked for it. And at the same time, Assistant itself — thanks to its apparent abandonment within Google — is also now flailing at those same tasks, which it once handled handily.

That adds up to create a lose-lose for everyone — except for maybe Google’s business department, which can now tell investors it’s pushing new boundaries and leading the way in generative AI development.

The most maddening part of all is that it didn’t have to be like this. Google could have thoughtfully worked out the best parts of Gemini, as an assistant, and then integrated those elements into the existing Assistant framework in a way that’d feel like an upgrade and expansion instead of a rug-swept-out-from-under-us degradation. It could have kept the system and the brand it spent years building up while simply sprinkling new capabilities in instead of doing its typical Google thing of pulling a complete 180, giving up on something entirely, and then leaving us — as its users — to sort out the mess.

More than anything, Google could have waited until Gemini was actually in a reasonably ready-for-prime state before rolling it out and pushing Android phone owners to switch over to it — while simultaneously dropping the ball on the existing Assistant platform and leaving us all in a lurch with no clear answers or direction.

All of this brings me back once more to the question I posed at the start of this year, when the effects of the tech industry’s hype-driven AI obsession were just starting to become apparent:

How much of the current rush to cram some form of “AI” into everything imaginable is actually about what’s useful and advantageous for us, as human users of these creations? And how much is more about chasing the latest buzzword du jour and finding a reason to use the term “AI,” no matter what it accomplishes or how it fits into our lives?

I’ve said it before, and I’ll say it again: Most of us don’t need a “creative” chatbot at our fingertips all day long, in every area of our Android experience. We don’t need on-demand image and text generation at our constant beck and call. And we certainly don’t need long-winded, on-screen answers of questionable accuracy for our short spoken questions.

What we need is a simple, reliable task-handler and an accurate and concise info-relayer. Assistant established the framework for that. And seeing Google throw that all away now and start over from scratch with Gemini — while forcing us to suffer along with its slog back toward a state of basic reliability — sure doesn’t feel like “progress.”

We may well reach a point where Gemini genuinely grows into a fully featured, reliable replacement for Google Assistant on Android and across the rest of Google’s ecosystem. I certainly hope so! But instead of bringing us to that point in a calm, carefully planned, and sensible-seeming manner, Google’s forcing us through a painfully rocky long-term transition — with a new tool that isn’t up to snuff, an old tool that’s being left to fall apart at the seams, and no meaningful guidance as to how all of this will ultimately play out.

That doesn’t seem like a smart way to handle things to me. And, I don’t know about you, but in this case, I don’t even need a half-baked virtual assistant to tell me why.

Get unmatched insight on all the news that matters with my free Android Intelligence newsletter. Three things to know and try in your inbox every Friday, straight from me to you.
Android, Google, Google Assistant, Voice Assistants   Read More Computerworld