Hardly, but I applaud the effort they do to remain in existence.

They might get relegated to niche use cases, or pivot like Midjourney. I'm glad they exist, even for the sake of fighting against the duopoly that's just forming with OpenAI and Anthropic.

Perplexity can fight Gemini for that 3rd spot. Gemini is kind of fumbling it, so they have a real fighting chance, imho.

▲

borg16 2 days ago | parent | next [-]

I do not understand how you can conclude that they both are even competitors - along any dimension

Perplexity has now become a speciality tool used for a few use cases primarily via third party models (unsure if they are still continuing to develop their own mdoels and where they rank)

Google on the other hand is still actively trying to develop SOTA models (or at least 3rd best) and also is widely used amongst cost sensitive projects.

I see no comparison between them worth putting them alongside each other.

▲

abirch 2 days ago | parent | prev | next [-]

Why is Gemini fumbling it? I use it for most of my personal projects and it's great for those.

▲

warkaree 2 days ago | parent | next [-]

I use Gemini it's bundled with my phone contract. Its privacy is nightmare there is no opt out.

▲

petcat 2 days ago | parent | next [-]

From and investor and "market" perspective, the only AI that matters is the one built into your phone, TV, car, and all your smarthome appliances. Google and Apple completely own this space.

If anything Anthropic and Open AI are at a severe disadvantage. Open AI at least provides backend model services for Siri. No idea what Anthropic does.

▲

spogbiper 2 days ago | parent | next [-]

> Open AI at least provides backend model services for Siri.

Do they? I thought that was Google/Gemini now.

https://www.cnbc.com/2026/01/12/apple-google-ai-siri-gemini....

	▲	jcomp 2 days ago \| parent [-]
		I sill have a ChatGPT extension option in Siri settings on 27.2; perhaps the default Apple Foundation model is Gemini-backed and the ChatGPT option still remains

▲

browningstreet 2 days ago | parent | prev | next [-]

Funny Friday

▲

ivanmontillam 2 days ago | parent | prev [-]

> From and investor and "market" perspective, he only AI that matters is the one built into your phone, TV, car, and all your smarthome appliances.

They can still be disqualified culturally speaking, as in, yeah every time you unbox a new smartphone it forces you to accept terms and conditions, yet you never open Gemini or Siri. That's great for the someone that will get that promotion and pay raise.

An investor sees that app installs are increasing and that's "great for business," yet it's shallow because the real app usage isn't that high.

"How come we have so many new installs but Claude is far more used?" They have all that telemetry, they should know that for certain.

Google and Apple can force installation and even terms and conditions agreement, but they cannot fake user commitment.

> Google and Apple completely own this space.

Microsoft bundled Internet Explorer, and once it became technically inferior (IE7+) to others, Chrome and Firefox ate the market share of it. Then it was Chrome all the way in.

Internet Explorer became "the browser you loved to hate," and they even did a campaign on that motto.

The technical debt was so heavy, they had to start over with Edge, and it still doesn't gain market share as it is expected. Just because it comes deeply ingrained in the mobile OS that still doesn't grant them ability to win the hearts of the users.

I'm one of those that delete the Gemini app every now and then (it becomes auto-installed with each OS update, it's hostile behavior).

Gemini and Siri still have to be good even if it's forced on our devices. We already know Siri lost it[0]. Siri is comically bad.

[0]: https://news.ycombinator.com/item?id=50104465

▲

0c3ca83 2 days ago | parent | prev | next [-]

Yeah, but all of the LLMs are a privacy nightmare, as is any Google product.

▲

jstummbillig 2 days ago | parent | prev [-]

Everything about that sounds like Google got it figured out for themselves.

▲

alterom 2 days ago | parent | prev [-]

I get better quality output from Perplexity than Gemini.

That said, I stick to Claude in the terminal for 99% of my LLM usage.

▲

abirch 2 days ago | parent | next [-]

Is this a gemini vs Antigravity-cli thing? I use agy as my cli at home and it's as good as Opus/Sonnet for me -- but I haven't tried very taxing tasks. I use Aragon/Flash. I'm not modifying large code bases, more like vibe coding throwaway projects at home.

▲

0c3ca83 2 days ago | parent | prev [-]

I too get better flavor from dog turds than rat turds.

▲

alterom 2 days ago | parent [-]

Not disputing this.

That said, LLMs can be fun toys regardless of the hype.

▲

0c3ca83 2 days ago | parent [-]

Toys, sure.

It's usually not a good idea to build gas pipelines directly to your toys in the middle of a climate crisis.

	▲	alterom 2 days ago \| parent [-]
		Again, not disputing this either.

▲

okdood64 2 days ago | parent | prev [-]

The CEO just needs to do more CNBC appearances and they're good.