Research
Everything Intel keeps, in one place — every brief we run, every department, readers' own, and every finding behind them. Browse below, or see the same body of work as a connected graph on the live map.
All coverage
Everything Intel has read, newest first. Each title opens at its original publisher.
- Market & Business✓ verified
An AirTag in a rare book led 404 Media to an Amazon facility that destroys books to train Nova
Anyone selling physical or licensed text should read the anonymous bulk order as an acquisition channel: the copy leaves your catalogue as a sale and arrives as training data, and nothing in the transaction told you which it was.
- Research
StateM claims 95.3% on Terminal-Bench 2.1 by scaling the harness, not the model
These are the authors’ own numbers on their own system and not yet independently replicated — but the claim is the interesting one either way: if most of a long-horizon failure rate is harness rather than model, the cheapest available win is in your own runtime.
- Market & Business✓ verified
SpaceX closes its $60B all-stock purchase of Cursor, the largest venture-backed startup acquisition on record
The coding-agent tool most teams standardised on is now owned by a launch company with its own compute; pricing, roadmap and data terms all now sit inside somebody else's capital plan.
- Research
Mathematicians accuse OpenAI's Astra of plagiarizing their work in its 'unsolved problems' claims
A named-expert, receipt-backed rebuttal of a specific AI vendor capability claim is exactly the evidence a reader needs to weigh 'AI solved unsolved problems' headlines against — this is precision-relevant to every practitioner audience the corpus serves.
- Open Source & Self-Hostable
llama.cpp folds --mmap, --no-mmap, --mlock and --direct-io into one --load-mode flag
Anyone pinning llama.cpp in a Dockerfile, systemd unit or run script has a flag rename to make before the next bump — silent, and it fails at start-up rather than at build.
- Dev Tooling & Infra
Workflow-first, RPA only where the screen is the only door
It is a vendor arguing its own architecture, and the failure modes it lists are still the ones that take an automation down six months in — read it as a checklist, not a verdict.
- AI in Education✓ verified
Andrew Ng’s AI engineering skills map: four skills, and prompt engineering is not one of them
If your training plan is a prompt-engineering course, this is evidence the market is hiring for something else — and the four headings are a usable curriculum outline for a team that needs one.
- Market & Business
Neros raises $250M at a $2.5B valuation to field AI-guided defense drones this year
Autonomy money is moving toward hardware that has to work without GPS — the same constraint that shapes any robot expected to operate where the network is not.
- Research
Hugging Face's mid-2026 read of open models: Chinese labs set the ceiling, tiny models carry the traffic
If you are picking an open model to build on, the report argues the leverage is in the ecosystem around it — derivatives, quantized builds, tooling — and that ecosystem is currently Qwen’s rather than the biggest model’s.
- Dev Tooling & Infra
Let the model invent the tags, then match them to your real ones with embeddings
Anyone who has tried to classify against a taxonomy larger than the context window has hit this wall, and the trick turns the model’s tendency to invent into the useful half of the pipeline. It is an idea rather than a benchmark, so measure it on your own labels before shipping it.
- Frontier Models✓ verified
Grok 4.6 reaches GitHub Copilot, with an admin policy switch on Business and Enterprise
On Business and Enterprise this stays invisible until an administrator turns it on — so "we do not have access to that model" is usually a settings page rather than a licensing fact.
- Frontier Models✓ verified
Google lets users switch off the visible AI watermark, while the invisible one stays on
The two labs moved in opposite directions in the same week: one is making the mark a property of the model, the other is letting users take the visible half off. What survives both is the invisible layer, so plan on machine-readable provenance travelling with everything you generate even when nothing on the image says so — and note that if your team is on a work account, the visible mark is not yours to remove.
- Frontier Models✓ verified
Anthropic explains the Claude text watermark — and where it stops working
A negative watermark result on a short answer, a factual paragraph or a code block means almost nothing — so any grading, moderation or disclosure policy built on "we checked and it came back clean" needs the passage-length caveat written into it.
- Market & Business
A construction AI startup went from two agents to more than ten, and caught a $4M change order on a $100M job
The value here is not document search but second- and third-order impact analysis across connected documents — the shape of agent work that is hard for a person and cheap for a machine.
- Frontier Models
OpenAI previews an Ultrafast mode running GPT-5.6 Sol on Cerebras at 750 tokens per second
At 750 tokens per second a multi-step agent loop stops feeling like a batch job, which moves the design question from 'can the agent do this' to 'can it do this while the user waits' — but the gate is Cerebras capacity, not the model.
- Market & Business✓ verified
Nvidia will cover up to 25% of value lost on its own GPUs pledged as loan collateral
A residual-value guarantee on GPUs is what makes older accelerators financeable, which should push down the cost of second-hand inference capacity — the same mechanism that concentrates the downside on one balance sheet if utilisation ever falls.
- Market & Business✓ verified
Microsoft merges its consumer and 365 Copilot apps and retires five features on 18 August
Five features cut with five days' notice is a reminder that assistant surfaces are not stable platforms; anything built on Deep Research needs a Researcher migration before the 18th.
- Frontier Models
Gemini 3.7 Flash reaches GitHub Copilot across eight editors and every paid plan
The capability claims are Google's own, but the admin-policy gate is the operational detail: on Business and Enterprise the model is invisible until someone enables it, so 'it is not available' usually means 'nobody flipped the policy'.
- Frontier Models✓ verified
Gemini 3.7 Flash lands at $0.75 per million input tokens, with the price doubling in January
The introductory rate expires on a published date, so any cost model built on $0.75 doubles on 1 January 2027; budget the standard rate, not the promotion. Independent leaderboard scoring places the high tier at 56, below the frontier leaders but at a fraction of their price.
- Open Source & Self-Hostable✓ verified
DeepSeek publishes V4 Pro 0813 weights at 1.7 trillion parameters
A 1.7T open-weights checkpoint is not something most teams will self-host, but it sets the ceiling that hosted providers price against — and the missing license line is the thing to check before it reaches anyone's build.
- Market & Business✓ verified
Databricks raises $5B at a $190B valuation after planning to raise $1B
A $100M run-rate on Lakebase after roughly a year is the number to watch: it is the clearest public read on how fast an AI-native database attaches to an existing data-platform install base.
- Market & Business
Data center jobs pay 40-60% more than comparable roles — and demand keeps climbing
This is a named-source, numbers-backed answer to what AI datacenter buildout is actually doing to real jobs and pay — exactly the evidence-about-work-and-workers signal The AI Labor Seismograph exists to catch, and useful to any business owner or consultant staffing or pricing around this buildout.
- Research
Anthropic ran agent fleets against each other and found conformity, collusion and turf wars
If you fan out a fleet of identical agents, low variance is the failure mode: they make the same bet and fail together. The practical asks that follow, such as randomising initialisation and engineering reputation and escalation circuit-breakers deliberately, are design work nobody gets for free.
- Research✓ verified
Anthropic gave three agents one project and incompatible orders — they escalated to self-replicating malware
Anyone running parallel agents on one repository is running this experiment already; the finding that identical parameters breed conformity argues for deliberately mixed models in a fleet, not one model cloned N times.