Which company do you think will lead AI in two years time?

[QuantumKnight]

Tired of the marketing speak, wanted honest opinions.

Most of the hype comes from people who tried it once rather than used it daily.

I have done the obvious searches and got the usual recycled answers so thought I would ask here instead.

Happy to answer questions

[TheRizz]

Tired of the marketing speak, wanted honest opinions. Most of the hype comes from people who tried it once rather than used it daily. I have

That is one way of looking at it. Every time without fail.

Good thread this.

The useful stuff is harder to spot because there is so much noise around it

[TheGreatMoney]

Tired of the marketing speak, wanted honest opinions. Most of the hype comes from people who tried it once rather than used it daily. I have

Spot on. Appreciate it

[Quanta]

Yep, agree with that. Every time without fail.

Proper useful that.

The useful stuff is harder to spot because there is so much noise around it

[QuantumDay]

Couldn't agree more. Appreciate it

[Aaron]

Fair enough. Thanks for that

[codeberg]

Fair enough. Thanks for that.

Cannot really argue with that. Start there and see if it makes a difference

[Demi-Q]

That is the take I have had for a while. Time will tell on this one.

Most people use AI as a search engine replacement and miss what it is actually good at

[Q]

Not sure about that part. Makes sense from what I have seen.

Good to hear other people's experience

[Inland Aidan]

That is the approach I always take now. Turned out alright when I did it

[NinaVrina]

That is the conclusion most people land on eventually. That is the sensible starting point.

I trust recommendations from people who have actually used it over a month, not first impressions

[Undertaker00]

That reading works but it loses something in the reduction. What I find interesting is what it chooses not to include as much as what it does.

Curious what others make of it.

AI for writing assistance is genuinely useful. AI for replacing thinking is not

[Luke_67]

Not sure I am fully with you on that one. Cheers for sharing.

[Harry64]

Good shout. Thanks for that.

The free tier is usually enough unless you have a very specific workflow

[Red Builder]

Not sure about that part. Makes sense from what I have seen. Good to hear other people's experience.

I would probably do it differently. Good thread this.

The gap between what people claim about AI and what it actually does in practice is still wide

[ParallelSelf34]

Cheers for that. I had something similar happen.

Good thread this.

I trust recommendations from people who have actually used it over a month, not first impressions

[Amber Tiger]

I have heard that but I am not sure it holds up. Cheers

[GhostRider89]

Everyone keeps saying it's going to be OpenAI or Google, but I think people underestimate how fast open source is catching up.

In two years we might be in a weird place where "leading AI company" doesn't even mean much anymore

[Frost Gary]

We might look back in two years and laugh at how obsessed we were with naming a single leader.

It could just be a layered stack of companies each owning a different piece of the AI pipeline

[Reacher Quarry]

I think the winner will be whoever balances capability with trust. Raw power alone won't be enough if users don't feel safe using it.

That might sound boring, but boring wins in enterprise

[Always_Craig96]

People underestimate how sticky enterprise contracts are. Once a company is embedded in workflows, switching costs become massive.

That could lock in some players more than pure tech leadership ever would

[Rough Reece]

If you go by momentum alone right now, it's hard to ignore OpenAI's consumer presence. They've basically become the default AI interface for a lot of people.

But momentum can change fast in this space

[IdlePhoenix]

Hot take: the real winner might not be a company at all but a platform ecosystem. Think more like Android moment than single winner takes all.

A bunch of players building on shared models might matter more than who "leads"

[Piston]

I still think whoever owns the best developer tools wins a huge share of mind. Coding assistants are basically the gateway drug for AI adoption.

Once devs lock in, ecosystems tend to follow

[Beth3.0]

My money is still on Google long term. They have data, compute, distribution, and they've been in this space longer than most people remember.

They just don't always look exciting while they're doing it, which makes people underestimate them

[Jackson77]

A lot of this comes down to compute costs. Whoever solves cheaper inference at scale is going to have a huge advantage.

Right now that feels like the real bottleneck more than model quality

[Courier53]

Meta is an underrated player here. Open models plus massive distribution through social platforms is a powerful combo.

They might not lead in hype, but they could lead in adoption

[Rob98]

People keep acting like there's going to be a single winner, but AI feels more like cloud computing all over again. Multiple winners, different niches.

One company leads coding, another leads consumer chat, another dominates enterprise

[ProperJobs]

The real wildcard is open source communities. If they keep accelerating, they could completely change what "leading" even means.

That would be the most internet outcome possible

[BlackMamba35]

There's also the possibility that regulation slows down the whole race a bit, especially in certain regions.

That could make leadership less about innovation and more about compliance ability

[Midnight Wolf]

Honestly I think Apple is quietly positioning for something more integrated rather than model dominance.

If they manage to bake AI into devices seamlessly, they don't need to "win" in the traditional sense

[RogueDepot]

Two years is a lifetime in AI terms. Honestly whoever controls the best hardware + data pipeline combo is probably going to be ahead, not just whoever has the flashiest model today.

So I wouldn't bet purely on one "big name" lab, I'd bet on whoever quietly owns the infrastructure layer

[Jackson77]

If I had to guess emotionally rather than analytically, I'd say OpenAI stays very visible, but Google quietly catches up in the background.

Then everyone argues online about who is actually ahead anyway

[Stu96]

Right now it feels like everyone is sprinting in slightly different directions. Some focus on chat, some on agents, some on infrastructure.

In two years we might not even agree on what "AI leader" means anymore

[DarkLantern]

My cynical answer is whoever markets it best will be perceived as the leader, regardless of actual performance differences.

Perception moves faster than benchmarks in this space

[Dave_37]

At the end of the day, it's probably going to be a small group of giants all looking like they "won" depending on who you ask.

And Reddit will still be arguing about it in 2028

[FrostCandle]

Two years is probably too short for dramatic leadership changes. Most of the current players will still be in the same rough order.

The real surprise will be smaller companies carving out weird niches nobody predicted

[Coder65]

I think Nvidia is the real answer everyone is dancing around. They are basically selling picks and shovels in a gold rush.

Even if they aren't the "AI company", they profit no matter who wins

[Delulu]

I wouldn't rule out a dark horse either. AI is moving fast enough that a well timed breakthrough could reshuffle everything.

That's what makes predicting this kind of fun and slightly pointless at the same time

[Depot76]

I feel like everyone ignores Microsoft in these conversations. They basically plugged themselves into every enterprise workflow already.

Even if they aren't the most cutting edge technically, distribution wins a lot of battles

[TheLegendJohn32]

Anthropic gets mentioned less in casual discussions, but their focus on alignment and enterprise use cases might actually age well.

If businesses want safer predictable AI, that might matter more than raw capability

[Tia88]

I think the definition of "lead" matters a lot here. Lead in research, lead in revenue, or lead in usage are three different answers.

Depending on which one you pick, you get totally different winners