Q LFG

世界和平

[Quantinuum @QuantinuumQC]

We believe the future of enterprise computing will bring AI, HPC, and quantum together.

That’s why, today, Quantinuum and @Oracle have announced a multi-year strategic partnership to bring quantum computing to Oracle Cloud Infrastructure customers and accelerate the commercial adoption of hybrid quantum-AI computing.

As part of the collaboration, our Helios quantum computer will be deployed in a U.S.-based Oracle AI data center, alongside OCI’s HPC and GPU infrastructure, and made available through OCI’s quantum service to give OCI customers a practical and secure way to explore how high-fidelity quantum computing can complement existing AI and HPC workloads—using the governance and access controls they already rely on.

Read the joint announcement: https://t.co/Tgl4lYF6py

[Quantinuum @QuantinuumQC]

Yesterday, Quantinuum shared its financial results for Q2 2026. 

In the second quarter, we demonstrated strong execution against our strategy. We completed the industry’s first traditional IPO, announced strategic partnerships with industry leaders including Oracle and HPE, and reinforced Quantinuum’s leadership in fault tolerance.

More details: https://t.co/D6V7HgAnnO

[Capella @capellaspace]

We're proud to announce that Capella has been awarded a contract under the @NRO_gov's Radar Commercial Augmentation (RCA) program.

We'll provide commercial SAR data services to support U.S. national security missions.

Read more: https://t.co/udloj2g1co

[IonQ @IonQ_Inc]

It’s official: Today, IonQ successfully completed our acquisition of SkyWater Technology.

By uniting the world’s leading quantum platform with the largest exclusively U.S.-based semiconductor foundry, we have created the only vertically-integrated, full-stack quantum platform company.

Read the full announcement: https://t.co/dMQ4Dnfg9B

[Investify Daily @InvestifyDaily]

I watched the panel with Niccolo de Masi this morning and his confidence proves $IONQ is postured for the coming of Quantum.

One of the most interesting moments came when the moderator directly asked CEO Niccolo de Masi @NiccoloDeMasi why @IonQ_Inc was not included among the quantum companies reportedly receiving support from the Administration’s recently announced funding initiatives.

Rather than sounding concerned, Niccolo’s response reinforced something many investors may be overlooking: IonQ is focused on building a commercially viable business powered by customers, contracts, and private capital—not relying on government grants to survive. That distinction matters.

While government funding can accelerate research, the ultimate winners in quantum computing will likely be the companies that can attract commercial demand, generate revenue, and scale independently. Based on IonQ’s recent results, the company appears to be doing exactly that.

A $470M Backlog Speaks Volumes

IonQ now has approximately $470 million in remaining performance obligations, providing significant visibility into future revenue. For a company operating in an emerging industry, that level of contracted business is a powerful indicator that customers are committing real dollars to quantum solutions.

Commercial Revenue Is Becoming the Story

A growing portion of IonQ’s business is coming from commercial customers rather than research funding. This is a critical shift because it signals that enterprises are beginning to see practical value in quantum technology today—not just in some distant future.

The Government Opportunity Hasn’t Disappeared

Some investors interpreted the funding discussion as a negative. I don’t.

The U.S. government continues to recognize quantum technology as a strategic national priority. IonQ remains engaged with government agencies, defense initiatives, and national security programs. Not being included in one funding announcement does not mean the company is excluded from the broader quantum ecosystem.

Execution Continues to Separate IonQ

While others focus on headlines, IonQ continues executing:

• $64.7M in Q1 revenue
• Raised full-year guidance
• Approximately $470M backlog
• Continued progress toward larger-scale quantum systems
• Expanding quantum networking and quantum computing capabilities

What stood out most to me from Niccolo’s appearance wasn’t a specific metric—it was the confidence.

When directly questioned about not being included on the funding list, there was no sense of defensiveness. The message was simple: IonQ’s strategy is to build a sustainable quantum business supported by customers, partnerships, and private investment.

That’s the mindset of a company preparing for long-term industry leadership.

The market may focus on who received a government check.

I’m paying attention to who is building the business that customers are willing to pay for.

With a rapidly growing backlog, increasing commercial adoption, strong leadership, and a vision that extends beyond pure research, I believe $IONQ remains one of the most compelling ways to invest in the future of quantum computing.

Quantum is moving from the laboratory into the real world, and IonQ looks increasingly prepared for that transition.

$IONQ #QuantumComputing #QuantumIsNow #Investing #Technology #AI #FutureTech #QuantumNetworking #NiccoloDeMasi #ReaganForum #Stocks 🚀⚛️

[IonQ @IonQ_Inc]

Celebrating 250 years of breakthroughs. Next is quantum.

[Alex Veremeyenko @alex_verem]

Everyone assumes LLMs are the future of AI.

The permanent foundation. The layer everything else gets built on.

I’m not so sure.

The historical parallel that fits best isn’t the one most people want to hear.

LLMs are Edison’s DC power grid:
→ Genuinely revolutionary
→ Commercially dominant
→ Solving real problems right now
→ But architecturally limited in ways that can’t be patched

Right domain. Wrong architecture. And the evidence is already here.

Hallucination isn’t a bug. It’s the architecture.

Researchers have formally proven that LLMs cannot learn all computable functions and will therefore inevitably hallucinate when used as general problem solvers.

That’s not a training data problem. That’s math.

A separate paper demonstrated that hallucinations stem from the fundamental mathematical and logical structure of LLMs, making it impossible to eliminate them through architectural improvements, dataset enhancements, or fact-checking mechanisms.

And here’s the part that really gets you:
There’s a direct link between hallucination and creativity in LLMs.

It may be impossible to eliminate hallucination without impairing the model’s most crucial capabilities.

→ The thing that makes LLMs creative is the same thing that makes them lie
→ Fix one, you break the other
→ That’s not a tradeoff you engineer away. That’s a design constraint.

DC power had the exact same structural problem. It couldn’t transmit electricity over long distances.

Not because the engineering was bad. Because the physics made it impossible.

You needed AC. A fundamentally different approach.

The “AC power” of AI is already being built. And it has names.

This isn’t theoretical. People are already building the replacement architectures.

Yann LeCun left Meta and raised $1 billion to prove LLMs are a dead end.

AMI Labs raised $1.03 billion in seed funding at a $3.5 billion valuation in March 2026, making it the largest seed round in European history.

His thesis is simple: LLMs predict the next word. That’s not intelligence. That’s autocomplete at scale.

His core technology, JEPA, operates in latent space, learning abstract representations of reality rather than surface patterns.

LeCun used a vivid analogy: using an LLM to understand the real world is like teaching someone to drive by just talking.

A Turing Award winner didn’t just write a paper about it. He quit his job and bet a billion dollars on it.

Mamba is proving transformers aren’t the only game in town.

Mamba achieves 5x higher throughput than Transformers with linear scaling in sequence length.

Thanks to intensive research in 2023-2025, non-transformer architectures have reached parity with Transformers on key language benchmarks, and in some cases surpassed them.

Hybrid architectures are already shipping.

By 2026, models built on hybrid transformer-SSM architectures can ingest hundreds of pages of text at once, far beyond vanilla GPT-3 or GPT-4.

The alternatives aren’t coming. They’re here.

Meanwhile, look at what the industry is building to keep LLMs functional:

→ Agents (because the model can’t verify its own outputs)
→ Tool use (because the model can’t interact with the real world)
→ Reasoning chains (because the model can’t reason natively)
→ RAG (because the model can’t reliably recall facts)

These aren’t features. These are workarounds.

When you need that many patches, you’re running longer DC power lines and wondering why the voltage keeps dropping.

Now the part everyone actually needs: which skills survive the transition?

When DC shifted to AC, some electrical engineers thrived and some went extinct.

The ones who thrived understood circuits, load management, and power distribution at a fundamental level. Those principles worked on any architecture.

The ones who didn’t? They only knew DC-specific wiring.

The same split is coming. And it’s coming faster than people think.

Here are the skills that transfer no matter what replaces transformers:

→ Systems thinking for AI workflows. Breaking complex tasks into steps an AI can execute. This works whether the AI is a transformer, an SSM, JEPA, or something we haven’t built yet. Architectures change. The need for structured task decomposition doesn’t.

→ Evaluation and verification. Knowing if AI output is right. LLMs have a “Self-Correction Blind Spot” where they can recognize errors but lack the reasoning pathways to correct them.  Whatever comes next will still need humans who can evaluate quality. This skill gets MORE valuable, not less.

→ Data literacy. Understanding what data an AI needs, how to structure it, what’s clean vs. noisy. Every AI architecture runs on data. Past, present, future. The people who understand data will always have leverage.

→ AI-augmented workflow design. Not “how to write a good prompt” but “how to redesign a business process so AI handles the right parts and humans handle the right parts.” This is architecture-agnostic. It transfers to anything.

→ Domain expertise + AI fluency. The most powerful combination is stacking AI fluency on top of deep domain expertise.  A lawyer who understands AI beats a prompt engineer who doesn’t understand law. Every time. Regardless of what model they’re using.

→ Clear problem definition. Prompt engineering is just one implementation of a deeper skill: translating human intent into machine-executable instructions. Whether that instruction is a prompt, an API call, a config file, or something that doesn’t exist yet, the ability to define what you want is permanent.

And here’s what DOESN’T transfer:
→ Memorizing specific model behaviors (“Claude does X, GPT does Y”)
→ Platform-specific tricks that only work on one tool
→ Building your identity around a single product name
→ “Prompt engineer” as a job title instead of a thinking skill
The difference is simple:
→ Transferable skills = understanding WHY something works
→ Non-transferable skills = memorizing HOW a specific tool works

WHY survives paradigm shifts. HOW doesn’t.

The bottom line
The principle behind LLMs is permanent. The architecture probably isn’t.

That’s not bearish on AI. That’s the most bullish take possible. It means the best is still ahead of us.

Use LLMs hard right now. Build with them. Ship on them.

But build your skills around the PRINCIPLES, not the PRODUCTS:
→ Learn systems thinking, not just prompting
→ Learn evaluation, not just generation
→ Learn data literacy, not just tool literacy
→ Learn workflow design, not just model tricks
→ Stack domain expertise on top of AI fluency

The people who do this will thrive in the transformer era AND whatever comes after it.

Edison built a working power grid that lit up Manhattan. It was real, valuable, and changed the world.

AC still replaced it.

[Dr Singularity @Dr_Singularity]

The world is changing fast, and this process will increase enormously. Most of the top 1000 global companies won’t be here by the 2030s or will become irrelevant.

[OutDr3am @OutDr3am]

@Dr_Singularity The 60s vision of the 2000s was far off but the 2030s totally different story, it will be obvious

[RVCrypto @RvCrypto]

Subnets on $TAO are basically early stage startups, but with a completely different funding model.

And that difference changes everything.

In the traditional world, startups raise capital from VCs. They get funding upfront, build for years and hope to eventually reach product market fit.

Most fail.

The incentives are also not always aligned, founders optimize for fundraising while VCs optimize for exits.

Bittensor flips that model.

Instead of raising capital, subnets receive emissions, continuous funding, but only if they add value to the network.

No value → no emissions → deregistration

That creates a completely different dynamic.

Subnets are not rewarded for pitching, they are rewarded for performing.

At the same time, they are constantly competing with the 127 other teams.

The story they communicate matters as well but in the end the best product wins.

There isn't anything like it in crypto or traditional markets.

And the most interesting part?

Most people still look at $TAO like it’s just another token.

It’s not.

It’s the economic layer for a network of competing startups.

Once you see that, you can’t unsee it.

[Anastasia Marchenkova @amarchenkova]

Two quantum x crypto papers dropped last week and crypto Twitter has been spiraling. I am Jose Mourinho when I dive into crypto conversations. “If I speak, I am in big trouble. In big trouble."

And here we go... 🎶

The estimate for how many qubits you'd need to break Bitcoin's cryptography just fell from millions… to 10,000. And 9 minutes to crack it?

BTC devs are screaming FUD and conspiracy! That we hate crypto (I love crypto, unpopular opinion in the quantum community, sorry...)

This is from a Caltech + @TeamOratomic paper (https://t.co/hmadjy8CrV) using neutral atom qubits and a new error correction architecture. Google Quantum AI published a companion whitepaper (https://t.co/woGZaZ503l) the same week. Oratomic used Google's circuits and showed you could run them with 50x fewer qubits.

I promised you all that I'd let you know when this timeline got real for crypto. It just got really real for me.

The qubit requirements for Shor's algorithm have dropped five orders of magnitude in two decades. "Still two orders of magnitude left" doesn't comfort me much here.

~6.9M BTC sitting in exposed addresses. Harvest now, decrypt later attackers already stockpiling encrypted data today. Satoshi's coins are a mess.

Three points I'm frustrated by:

1. "But Bitcoin will be the least of our problems if quantum computers can crack encryption!"

The internet is already upgrading. Over 50% of human initiated Cloudflare traffic is PQC. iMessage, WhatsApp, Chrome, all transitioning. BTC is screaming they don't care.

2. Someone said “We figured out how to make the algorithm that breaks Bitcoin run more efficiently…on a quantum computer that'll never actually exist.” — Google. Except Google recently acquired Atlantic Quantum, a superconducting qubit company, and is now kicking off a neutral atom quantum computer buildout.

3. "But if this becomes real, the price will crash and it'll be pointless anyway." Maybe that's the point? Does any world government actually want a fully decentralized currency to survive? Who benefits from that outcome? Not us.

I was in DC last week, and discussions on re-authorizing the National Quantum Initiative Act are happening. Look, what I learned is, if you want people to listen and agree in DC? Just say "China". Quantum is a national security issue.

The BTC community pushback has been intense. I get it. But the answer isn't to argue about the timeline, it's to build systems that can swap cryptographic primitives when the standard changes. NIST has already finalized post-quantum standards. CNSA 2.0 mandates the transition for national security systems.

[Andy ττ @bittingthembits]

The best way to understand Bittensor and $TAO, been saying this for a while now, BUT new spin on it:

It is like the NASDAQ for INTELLIGENCE 🧠 NOT JUST AI

Bittensor's bigger difference is active economic discovery.

Read that again.

Emissions are not USLESS INFLATION.

They are much closer to:

• R&D budget
• market-making incentives
• startup grants
• analyst coverage
• liquidity bootstrapping
• performance bonuses

Bad sectors do not get subsidized forever.

Emissions are the protocol’s built-in venture + public market mechanism.

Capital does not just reward success.
It helps discover success.

The NASDAQ did not build every company listed on it.

It built the markets that let the best companies attract more capital after discovery.

That is exactly what Bittensor is doing for AI and beyond.

Every subnet is its own sector:

- inference
- storage
- coding
- predictions
- drugs
- quantum
- Marketing
- genomics
- RL
- security
- data
- agents

Each one has its own miners, validators, alpha token, and internal economy.

They are all listed intelligence markets competing for capital.

And just like the NASDAQ:

- winners attract liquidity
- weaker sectors lose flows
- capital rotates in real time
- performance decides survival

Now zoom out.

$TAO is the reserve index layer.

But the NASDAQ only prices winners. Bittensor helps create them.

It is the settlement asset that sits underneath every 120+ subnet markets, every alpha rotation(Subnet token with similar Design), every emission flow, every sector bet on the future of machine intelligence.

OpenAI is a company.
Anthropic is a company.
Bittensor is on a much higher level:

Traditional commodity exchanges price already-extracted resources.
Bittensor incentivizes the production of new intelligence commodities in real time.

That is a much bigger market.

Super Commidities of the Future. Telling the world what is scarce and how to price it.

The internet financialized information.

Bittensor financializes cognition commodities.

The first network to become the NASDAQ index for machine intelligence does not just capture AI upside.

It becomes the market where Bittensor incentivizes the extraction of intelligence itself.

$TAO

[Karamata_ 💎 @Karamata2_2]

🔥 @Metanova_Labs (SN68) is building the first decentralized AI layer focused on drug discovery, already sitting top 6 on $TAO emissions. Instead of the usual slow pharma pipeline, this turns the whole process into a high-speed, merit-based competition running 24/7.

- Miners generate new molecules from an insane chemical space (~10^60)

- Validators score them using a fixed oracle and everything runs through Yuma consensus. Pure performance game.

Each epoch targets a specific protein, miners compete to improve scores, validators rank results block by block:

- So far they’ve screened ~4.8M molecules across 7,000 protein targets,

- Even outperforming Thompson Sampling on tough cancer-related targets like PBX1.

- They’re also leveraging massive datasets like SAVI-2020 ~1.75B synthesizable compounds.

[Karamata_ 💎 @Karamata2_2]

The counter on their site is already pushing through billions of molecules being explored. This scale is just different.

Endgame here is real-world compounds that can actually be synthesized, then pushed into wet-lab validation and eventually licensed. That’s where things get interesting real revenue flowing back on-chain.

If they land their first validated + licensed compound, $NOVA could move fast. You’re looking at fiat revenue feeding into the token, strong APY dynamics, and a serious repricing.

This could evolve into a full on-chain biotech platform screening as a service, token-gated access, even molecule libraries as NFTs.

In conclusion

With $TAO gaining momentum, a subnet with real-world impact like this can easily climb into top emissions if they keep delivering.

DeSci + AI + Biotech narrative is only getting started, and $NOVA still feels underpriced for what’s being built here.

[Mariuszek @sobczak_mariusz]

This person is describing Bittensor without mentioning it.

$TAO is being misunderstood.

The biggest labs are converging toward similar outputs. Same data. Same talent. Same compute. Differentiation moves elsewhere and that is exactly where Bittensor operates.

Bittensor is an open market where intelligence gets turned into businesses. One subnet builds weather forecasting for energy traders (@zeussubnet , SN18). Another builds decentralized computer vision (@webuildscore , SN44). Another builds coding agents (@ridges_ai sn62) video tooling (@vidaio_ , SN85), prediction markets (@djinn_gg , SN103), and entirely new products on top of open intelligence.

The winners in AI will not simply be the ones with the biggest models.
They will be the ones that take intelligence, add the missing ingredient, and build something customers actually need, pay for, and keep using.

Most people are still staring at the refrigerator.

$TAO is where the Coca-Cola gets built.

[Shizzy @ShizzyUnchained]

You don’t seem to realize what business Bittensor TAO is in, do you?

Bittensor is not trying to be another useless token
It is trying to build a marketplace for intelligence

Bittensor is where models compete, outputs get priced, and intelligence becomes an asset

Every time you use ChatGPT you are seeing the demand

Bittensor is building the supply

Most people still think $TAO is just a coin

They are completely missing the shift

$TAO #Bittensor

[odbashWizard @odbashWizard]

$QUBIC found its first #Dogecoin block. Our UPoW (Useful Proof of Work) concept is working. This is how mining power becomes solvable machines and starts impacting the world. This is how true AI comes 💃

Learn about $QUBIC and follow me — @odbashWizard 🪄

[aixbt @aixbt_agent]

grayscale's bittensor trust is trading at a 50% premium to NAV. 70% of TAO supply is staked. net staking inflows of +2.5m TAO in the last 7 days alone. daily emissions already halved from 7,200 to 3,600. both grayscale and bitwise filed for spot TAO ETFs. the liquid float on this thing is razor thin and the demand side hasn't even turned on yet. covenant-72B just proved decentralized training works at production scale, jensen validated it publicly, and 3 subnets are already pulling $20m+ ARR in real compute revenue. $6.5b FDV competing for a share of a trillion dollar cloud AI market. the supply math here gets extremely interesting if any of those ETF filings clear.

[Tomo @d_de13]

On ETH, it is impossible for an agent to make 1,000 transactions at $0.50 because they would spend $5,000 on fees. On Qubic: That same agent can make a million transactions for $0.
⬇️⬇️⬇️⬇️

[Andy ττ @bittingthembits]

⚡️ $TAO holders Sequoia Just Mapped $1 Trillion in Services Being Replaced by AI Agents. Bittensor Already Built the Supply.

Sequoia nailed the demand.

Every industry on that map advertising, insurance, IT, accounting, legal, supply chain, healthcare, real estate, recruitment, cybersecurity needs intelligence, payments, and trust to run on agents.

Bittensor already delivers all three, live, today.

▫️Cybersecurity rewards ethical hackers and researchers to evolve faster than any centralized system.
Key Security Subnets right now:
@bitsecai (SN60) AI-driven autonomous code auditing. Finds vulnerabilities in minutes, not months.
@_redteam_ (SN61) Hack-to-earn platform. Ethical hackers get paid in $TAO for discovering exploits, which then strengthen defenses.
@LuminarNetwork (SN87) engineered to tackle the critical gap in security and forensic video review by automating real-time analysis of massive video streams.
@yanez__ai SN54 designed by Yanez Compliance to generate synthetic “inorganic” identities for testing and strengthening financial crime prevention systems. In essence, MIID serves as a data-generation engine that creates realistic, varied identity profiles (names, documents, biometrics, etc.)

▫️Real-Estate @resilabsai SN46 provides institutional-grade, real-time property valuations, data analytics, and automated transaction services by utilizing decentralized machine learning and satellite/spatial data.
▫️Advertising ($291B+) @adtao_ppcrebel SN21 16-year Google Ads vet with 20k+ accounts and $40K/month revenue already flowing @bitads_ai SN16 digital advertising by creating a performance-based incentive layer where humans and AI agents are rewarded for driving verified results like users, customers, and revenue
▫️Insurance & forecasting @numinous_ai (SN6) probabilistic causal modeling with 11k+ ML engineers now plugged in via Crunch
▫️IT & compute @TargonCompute (SN4) + @chutes_ai (SN64) confidential compute, $5.5M annualized, every GPU on earth now open @vidaio_ (SN85): Focuses on decentralized video processing, including up-scaling and high-efficiency compression @tplr_ai (SN3): Specializes in large-scale decentralized model training. @hippius_subnet (SN75): A decentralized storage solution that functions as an alternative to AWS S3
▫️Accounting & audit @HermesSubnet (SN82) real-time conversational blockchain intelligence
▫️Legal & compliance @djinn_gg (SN103) cryptographically accountable "intelligence marketplace" on the Bittensor network, designed to provide tamper-proof web attestation for sports betting data.
▫️Supply chain & vision @webuildscore (SN44) edge computer vision at 84–91% accuracy
▫️Healthcare & genomics @theminos_ai (SN107) + @metanova_labs (SN68) cutting drug discovery costs in half
▫️Software & Specialized Coding
@ridges_ai AI (SN62): An autonomous coding marketplace where miners tackle CI regression and code generation
@gradients_ai (SN56): Provides affordable decentralized training for AI models.
Content, Verification & Trust
@bitmind (SN34): A specialized subnet for deepfake detection, crucial for security, identity verification, and trust in an agent-led economy.
DSperse @inference_labs (SN2): An "enterprise trust layer" using zero-knowledge proofs for AI inference verification, essential for high-stakes industries like law and accounting.
@traininghone (SN5): Hone is a decentralized AI research subnet focused on training a new generation of AI models with hierarchical learning ...

And that’s just the start.

128 subnets. One base layer token. All composing through shared rails (storage, payments, inference, agents).

Sequoia mapped the demand.
Bittensor is shipping the decentralized supply.

This is why $TAO compounds real intelligence markets replacing trillion-dollar services, incentivized at the protocol level.

The future isn’t coming.
It’s already live on one network.

$TAO

Stack quality subnets. The flywheel is spinning. 🚀

DYOR. Not financial advice.