Ciente > Blog > News

Is SoftBank Doubling Down on OpenAI? Sells $5.8bn NVIDIA Stakes

Category: News

Is SoftBank Doubling Down on OpenAI? Sells $5.8bn NVIDIA Stakes
News

Is SoftBank Doubling Down on OpenAI? Sells $5.8bn NVIDIA Stakes

Is SoftBank Doubling Down on OpenAI? Sells $5.8bn NVIDIA Stakes

Softbank’s sudden exit from NVIDIA isn’t the financial giant cleaning its hands of AI. It’s merely revamping its investment strategies.

There has been rampant speculation (and sure-shot statements) about the AI bubble. But only a few in the market have really gauged if AI is truly creating a bubble, or it’s merely a precaution, a way of being cautious after the dot-com bubble.

Each different quarter this year has flagged down a warning that this potential bubble could burst. And send all of us toppling- the G7 have no long-term, sustainable AI model in sight. But they continue to invest billions of dollars into artificial intelligence.

Hearing of Softbank selling its stakes in NVIDIA for $5.8bn sent shivers down the market.

But it isn’t giving up on AI just yet. The financial institution is merely doubling down on OpenAI, which it believes holds more promise. Especially after it reported $15.99 billion in valuation gains driven by its OpenAI holdings.

Why?

For Softbank, its investment in OpenAI is more substantial. It’s also freeing up more assets to further invest in new avenues. And diversify its portfolio after they made a $30bn investment in OpenAI.

According to a few market analysts, it’s not as if Softbank is abandoning the AI route. It is still the shiny, glossy plaything. Only that the finance giant could have found newer toys, which it believes hold more potential than NVIDIA.

“Investors typically sell out of positions when they believe the valuation is too rich, the growth prospects for the company are less attractive than before, or they’ve found something better to back and need cash to make that investment,” chimed in AJ Bell’s investment director.

And further on, attempted to justify this sudden shift–

“Nvidia’s role in an AI world is already well known, yet OpenAI’s position is still evolving, so it might simply be that SoftBank sees the latter as a better way of profiting from the tech explosion going forward, rather than sticking with yesterday’s trailblazer.”

Honestly, there’s no stopping these institutions. So much so that the US’s economic model is now basically these seven giants sending a trillion dollars back and forth to each other.

image 11

Wired announcing that AI is the Bubble to Burst Them All, to Fortune declaring that “a collapse is definitely a possibility.” There are plenty of best-case scenarios that the tech leaders and investors are drumming up for you. And some are rosier than others.

But there’s no doubt when we say that all of these companies are tied together in financial deals that are ticking time bombs. Even the Big Short- Michael Burry, who rightly predicted the housing bubble of 2008, placed $1.1bn bets against NVIDIA and Palantir.

Isn’t this a strategic move, or is it all left to fate’s hands?

AI can be persuasive. But it’s all a gigantic mess, not merely a bubble.

Onto Project Suncatcher Could Space Be The Answer to AIs Latent Potential
News

Onto Project Suncatcher: Could Space Be The Answer to AI’s Latent Potential?

Onto Project Suncatcher: Could Space Be The Answer to AI’s Latent Potential?

Google, in alliance with the company Planet, hopes to launch its first couple of solar-powered satellite prototypes for Project Suncatcher by 2027.

The market is being driven distraught- is the AI bubble finally going to burst? And amidst this frenzy, maybe, just maybe, there’s an answer to all the speculations regarding AI’s true potential.

Since its escalating adoption, the world assumed that we might never decode the modern tech’s maximum potential in our lifetime. Well, we were wrong to presume that.

As the AI race unfolds right in front of us, we are left dumbstruck by Google’s moonshot plan. And while it sounds like the normal next step, it truly is a fascinating step towards opening new avenues for AI-led innovation. And gauge- is sky truly the limit for artificial intelligence?

Google’s Project Suncatcher is understood to be a moonshot research project. The crux? It’s taking AI to space. And honestly, the whole antic sounds exciting on the surface.

The success of this project could push companies to scale ML in space. And this would be powered by the most substantial energy source, our Sun. So, basically, the research project plans to assess whether a constellation of solar power-backed satellites, equipped with Google’s TPUs, can be connected by free-space optical links.

Google calls it the “future space-based, highly scalable future AI infrastructure design.”

And it’s already taking baby steps toward transforming this project into a reality. The first is creating modular designs for the small, interconnected satellites.

If the tech powerhouse figures this out, there’s a bright road ahead. A future where AI relies less on terrestrial resources. And more on the never-ending energy backup- solar power. These models would continuously churn out electricity. And facilitate eight times the productivity of current data centers.

However, there’s more to the story.

The resource-hungry data centers built on space? Could this be what the world needs right now?

Yes. If this is the missing puzzle piece to binding AI and sustainability.

These space-based AI data centers could harness the Sun’s clean energy around the clock. It could dispel the havoc created by the earth-bound data centers.

They’re thirsty for freshwater (but not for salt water, because wouldn’t that make things easier?). They’re driving up the utility bills. And a magnanimous demand for electricity in the surrounding areas.

On Earth, AI and sustainability can’t work in tandem. At least, not yet.

In response, Google’s Senior Director for Paradigms of Intelligence asserts, “In the future, space may be the best place to scale AI compute.”

But for now, Project Suncatcher remains an ambitious research project.

Meta Plans to invest $600bn in the US in Capex
News

Meta Plans to invest $600bn in the US in Capex

Meta Plans to invest $600bn in the US in Capex

Meta has recently announced that they would be investing $600bn in AI data centers by 2028. This is a bit misleading for many reasons.

Reuters recently covered that Meta will be investing around $600bn in the US via AI infrastructure in the next three years. This includes jobs and data centers, as well as other auxiliary infrastructure needed to grow their division.

This has been misconstrued by a lot of media giants. Some think that Meta is raising or investing $600bn in direct cash. No, this is capex.

Meta does not have $600bn. What they are instead promising is, through their initiatives, to create a system that injects that value of that amount into the US economy.

The key here is that through relatively small investments into its own AI projects, Meta will create economic incentives worth $600bn. But the question is: can they make do on their promise?

The AI bubble.

Yes, let’s talk about the AI bubble. The circular economy has been haunting the world for quite some time. The top AI companies have been under fire for allegedly moving money within their own ranks.

And the fear of the AI bubble rises among investors and stakeholders- including employees. Then there’s the doomsayers, scientists among them, warning of the emergence of superintelligence.

They make it sound like science fiction, but a growing school of thought believes not. This change in technology doesn’t just herald a change in our economic systems but also the way of life of many communities.

It is a disruption on a scale not known to our economies.

Based on these fears, apprehensions, and misinformation, do organizations know what they are doing with this tech?

Let’s hope that they do because the alternative is a scary one.

Apple to Revamp Siri with Google's 1.3 Trillion Parameter AI Model
News

Apple to Revamp Siri with Google’s 1.3 Trillion Parameter AI Model

Apple to Revamp Siri with Google’s 1.3 Trillion Parameter AI Model

Apple is finalizing a deal to leverage Google’s 1.2-trillion-parameter Gemini model to power the much-awaited Siri upgrade.

The thing about Apple’s famously self-reliant ecosystem? Sometimes it needs to call in the neighbors for help. And by “help,” I mean a cool $1 billion per year to rent Google’s brain.

For context, that’s roughly eight times more complex than Apple’s current 150-billion-parameter models. Eight times. Let that sink. And remember, every time Siri confidently misunderstood your simple request.

There’s an irony hidden here.

This is the same company that built its brand on doing things “the Apple way.” Privacy and control. The same company that’s been promising us a smarter Siri since, well, since Siri became a punchline. And now? They’re running Google’s models behind the scenes while marketing it all as Apple technology.

Honestly, Apple did test alternatives- OpenAI’s ChatGPT and Anthropic’s Claude before choosing Google. But here’s where it gets interesting: what reportedly tipped the scales wasn’t superior performance, but price. Nothing says “innovation leader” quite like shopping for the budget AI option.

Apple insists this is temporary.

The company claims it’s developing its own 1-trillion-parameter model that could be ready as soon as 2026. Sound familiar? It’s the same playbook they used with maps, weather data, and chips. Lean on someone else until you catch up.

Except this time, there’s no guarantee users will embrace the new Siri or that it can undo years of damage to the brand.

The planned spring 2026 launch gives Apple just enough time to slap its design language on Google’s tech and call it revolutionary. Gemini will handle the summarizing and planning functions- the parts that actually require intelligence.

Meanwhile, Apple already pays Google around $20 billion annually to be the default iPhone search engine. Now add another billion for AI assistance.

At this rate, Google isn’t just inside Apple’s walled garden. It’s paying the mortgage.

Tencent AI Joins the Race, Hopes to Rival OpenAI's Sora
News

Tencent AI Joins the Race, Hopes to Rival OpenAI’s Sora

Tencent AI Joins the Race, Hopes to Rival OpenAI’s Sora

A Tencent AI veteran just raised $50 million to build a rival to OpenAI’s Sora. The move exposes both Tencent’s quiet AI progress and the cracks within its innovation machine.

A scientist who used to spearhead the development of Tencent’s flagship AI model has just helped raise US$50 million for a new startup- one that’s positioning itself as a rival not just to Tencent, but to OpenAI’s video-AI product Sora.

At first blush: bold move. It suggests Tencent’s AI talent isn’t simply locked up in the mothership. And it means the “old guard” inside Chinese tech is saying: maybe the next wave isn’t from the giants alone.

But let’s dig into the wrinkle: If Tencent nurtured that talent, why is the spin-out happening now? Because the terrain of generative AI, especially video, is shifting fast. According to external reports, Tencent claims its internal AI platform already boosted R&D efficiency by more than 20%, with one system finding a quarter of code-bugs via AI.

In other words, internally, Tencent is advancing. Externally, though, you have a hungry startup creeping in with seed capital and a stated mission: make video creation as intuitive as chatting with GPT. That’s a sign the “comfortable space” inside Tencent may feel too slow or too conservative for some innovators.

Now the critical arc: Should we worry that Tencent’s own AI stack (including its foundational model “Hunyuan” and others) is losing its edge, or is this spin-out simply a sign of Warren-Buffet-style diversification? The $50 m is modest compared to Tencent’s scale; it’s not a moon-shot yet. Also, the startup is still early- even described as “year-old”.

What I’m leaning toward: This reflects two truths.

  1. Tencent knows it’s in a race- pressure from the US & domestic competitors. So, letting talent spill out (even indirectly) may be a strategic gesture: “We own the ecosystem whether you stay inside or spin-out”.
  2. Talent exits and seed spin-outs often signal internal friction: speed vs scale, risk appetite vs bureaucracy. For you, writing the piece, that tension is rich.

The verdict: Keep eyes on how the startup executes (video-AI is still messy). But watch Tencent: if one of its own goes rogue and wins big, it will expose blind spots in big tech’s innovation machine. For now, $50 m is a shot across the bow- not yet full broadside.

OpenAI Signs $38bn Cloud Computing Deal with Amazon.
News

OpenAI Signs $38bn Cloud Computing Deal with Amazon.

OpenAI Signs $38bn Cloud Computing Deal with Amazon.

OpenAI’s deal with AWS cements Amazon as the AI era’s infrastructure kingmaker. But also exposes how dangerously centralized and power-hungry the race for intelligence has become.

So here’s the thing: OpenAI has signed a $38 billion deal to use Amazon’s infrastructure. Yes, billions- granting them access to AWS datacentres and hundreds of thousands of Nvidia chips.

At first glance, this is the kind of muscle move that says, “We mean business in AI.” But dig a little, and you see something both bold and a little worrisome.

Bold because scaling frontier AI does, in fact, demand massive, reliable compute. OpenAI’s own CEO says this partnership “strengthens the broad compute ecosystem that will power this next era.”

Good, push the bounds, build the backbone. But what about the “worrying” part?

OpenAI simultaneously says it’s committed to 30 gigawatts of computing resources, enough to power about 25 million U.S. homes.

Now, compare that to revenue: OpenAI reportedly made around $13 billion annually (publicly, at least), yet has committed to a $1.4 trillion infrastructure binge.

Let the imbalance sink in. If you’re backing an AI war-machine, you’d better have a war budget, or the cash-flow won’t hold.

And then there’s Amazon. By taking OpenAI on this deal, Amazon becomes essentially the backbone- the pipes, the powerhouse. AWS is now deeply entwined with one of the most ambitious AI players. That’s smart for Amazon, no doubt. But for the broader ecosystem? This centralization raises vital questions about power, risk, and lock-in.

In short, OpenAI’s move is ambitious and deserves respect. But it may also be placing a staggering bet on a future where compute equals dominance. And AWS? They’re playing the infrastructure kingmaker. The risk is not just for the companies, but for the tech ecosystem:

When one deal holds this much sway, who watches the watcher?