Every capability we track

Yes. Modern robot vacuums map your home, avoid obstacles, mop, empty themselves and run for weeks without attention. They still cannot pick up the clutter first, and stairs remain their nemesis.

Yes, easily, from a very short sample — which is precisely the problem. Voice cloning is cheap, fast and convincing enough for audiobooks, dubbing and accessibility, and equally convincing to a relative receiving a scam call.

Yes, and it is genuinely good. Phones, earbuds and some glasses translate spoken conversation with a second or two of lag across major language pairs. It is excellent for travel and everyday chat, and still risky for legal, medical or contractual conversations.

Yes, mostly. Consumer phones and cheap call-screening services will pick up, transcribe and hold a short conversation with a caller, and they handle spam and simple appointment calls well. Anything nuanced — an emotional caller, a complex booking, a call that needs your judgement — still gets handed back to you.

Close. Assistants triage, summarise, draft in your voice and handle scheduling well enough that many people no longer read every message. Sending important replies unsupervised is still a bad idea, and quality drops on unfamiliar threads.

Very nearly. Low-earth-orbit satellite broadband delivers genuinely fast, low-latency connectivity almost everywhere, including at sea and in the air. Cost, terminal size and a handful of restricted countries are the remaining gaps.

It can write most of it, but shipping is still a human responsibility. Coding agents build features, fix bugs and open pull requests unaided, and teams genuinely rely on them. Architecture decisions, subtle security issues and long-lived maintenance still need engineers in the loop.

Yes for practice, explanation and patient repetition — with adult oversight. AI tutors adapt to a pupil's level and never tire. They are weaker at motivation, assessment integrity and knowing when a child is quietly disengaging.

It can remember a lot, not everything. Mainstream assistants now keep persistent memory of your preferences, projects and past conversations, and it noticeably improves them. Recall is still selective, occasionally wrong, and confined to whichever assistant you told.

For small, conventional apps, genuinely yes — a non-developer can now describe a tool and get something deployed that works. The gap appears later: security review, awkward edge cases, data modelling and the unglamorous maintenance that decides whether software survives its first hundred users.

For short clips, adverts and b-roll, yes — and it is already in commercial use. Sustained narrative video with consistent characters, dialogue and continuity across minutes remains difficult and needs heavy human editing.

It can run the routine layer. Scheduling, invoicing chase-ups, customer replies, bookkeeping categorisation and marketing drafts are all realistically automatable now. Judgement calls, exceptions and anything with legal weight still need an owner.

Partially, and only for specific things. Wearables have credible, regulator-cleared features for irregular heart rhythm, sleep apnoea signals and fall detection. Broad early warning of infection or disease is research-grade signal, not diagnosis.

It can react intelligently; genuine anticipation is patchy. Modern smart homes handle presence, light, heat and routines well, and LLM-driven assistants finally understand loose natural instructions. Predicting what you want before you ask still produces as many annoyances as wins.

It plans brilliantly and books adequately. Agents will research destinations, build itineraries and complete simple bookings. Multi-leg trips with dependencies, loyalty programmes and awkward change policies still go wrong in expensive ways.

Yes for everyday objects, landmarks, menus and text. Consumer camera glasses answer 'what am I looking at?' surprisingly well. Precision tasks — identifying a specific component, a mushroom, a medication or a face — are either unreliable, restricted or both.

For texting, increasingly yes. Unmodified handsets on supported networks can send messages and emergency alerts via satellite where there is no mast, which is a genuinely meaningful safety change. Satellite video calls from a normal phone are not what this is.

For travel and single-user focus work, yes. Display glasses give you a large virtual screen from a laptop or phone, and headsets do multi-window work convincingly. Comfort over a full day, text sharpness and price still stop most people ditching a desk monitor.

You can send a note-taker; you cannot credibly send a stand-in. AI meeting assistants join, transcribe, summarise and assign actions extremely well. Photorealistic avatars that speak for you exist, but participating convincingly in live discussion is beyond them — and socially fraught.

Partly. An assistant can plan meals, build a basket and reorder your usual items, and some retailers offer their own AI basket builders. Fully hands-off shopping — choosing substitutions the way you would, spotting a bad price, adapting to what is actually in stock — still needs you to check the basket before checkout.

It works often enough to be worth it, but not predictably. Services and agents will call or chat with providers, ask for retention offers and cancel subscriptions. Outcomes depend heavily on the provider, and some actively block automated agents.

You can buy one, but expect a companion rather than an assistant. Pendants, pins and earbuds will listen on demand, transcribe your day and answer questions well enough to be useful. Continuous all-day capture that reliably surfaces the right thing at the right moment, on a battery that lasts, is still beyond them.

AI supports diagnosis; it does not own it. Cleared tools read scans, flag findings and triage in clinical settings, and general models are startlingly good at plausible differentials. Regulation, liability and error modes mean a clinician signs off, and a chatbot is not a doctor.

In specific towns and suburbs, yes — commercial drone delivery is real and often faster than a van. Coverage is the problem: airspace approvals, noise, weather and payload limits keep it to a small fraction of addresses.

For simple, well-lit, well-known tasks, they can help. Camera-plus-assistant glasses will identify what you're looking at, translate a sign or talk you through a recipe. Step-by-step guidance through unfamiliar physical work — repairs, wiring, anything with consequences — still lags and misreads too often to trust.

In a growing number of towns and campuses, yes — pavement robots genuinely deliver groceries and takeaways, and they mostly work. Coverage is the catch: unless you live inside a served zone with tolerable pavements, it isn't an option, and remote operators still take over more often than the marketing suggests.

It can fill the basket; you should still press the button. Agents research products, compare prices and complete simple reorders competently. Letting one spend freely without confirmation invites the failure mode nobody wants: a confidently wrong purchase, at speed, with your card.

Only if you live in the right city. Driverless robotaxis run commercial services in a handful of mapped urban areas and genuinely work there. A privately owned car that drives you door-to-door, everywhere, in all weather, with nobody watching the road is not something you can buy today.

Yes — printed homes are occupied in several countries. What is printed is usually the concrete wall structure; foundations, roof, services and fit-out remain conventional. Cost savings are real but smaller than headlines suggest.

No, not reliably. Detectors work reasonably on known generators and degrade badly on new ones, on compressed social media re-uploads and on partially edited media. Content provenance signing is the more promising route, and adoption is still partial.

Not to a standard you should treat clinically. Wearables are good at heart rhythm, sleep staging and trends, and irregular-rhythm alerts have genuinely sent people to a doctor in time. Cuffless blood pressure is approximate and drifts; needle-free glucose remains unsolved despite two decades of confident announcements.

On specific motorway corridors, in permissive jurisdictions, yes — driver-out freight runs are happening commercially. Everywhere else, and for the messy first and last miles around depots and city streets, a human is still in the cab.

Yes, if you are extremely wealthy. Suborbital hops and private orbital missions genuinely fly paying customers. It is not tourism in any normal sense: seats are scarce, six to eight figures, and manifests are booked years out.

Barely, and this is becoming the important question. As agents log into services and spend money, sites need to know which agent is acting, for whom, and within what mandate. The standards exist in draft; the deployed reality is mostly API keys and hope.

Barely. A small number of countries have approved cultivated meat products, sold in limited restaurants or trial retail. Production volumes are tiny and costs remain far above conventional meat, so this is a novelty rather than a supply chain.

In research, users genuinely feel touch through nerve-interfaced prosthetics, and it measurably improves control. Commercially available advanced prosthetics offer excellent grip patterns but only limited or no true sensory feedback, and cost is a major barrier.

In pilots, under supervision, on narrow tasks. Several manufacturers have humanoids moving totes and feeding lines in real facilities, which is a genuine step up from stage demos. Throughput, uptime and cost per pick are still nowhere near the wheeled robots and fixed automation already doing this work.

For contrived benchmarks, sometimes; for work you would actually pay to have done, not yet. Error rates remain the wall. Error correction has improved enough that the roadmap looks credible rather than mystical, but commercially valuable quantum advantage has not arrived.

In a lab, yes. In your house, no. Research and demo robots fold towels and shirts at slow speed under good lighting, and general-purpose humanoids show it off constantly. Nothing you can buy will reliably empty a real basket of tangled mixed laundry.

For a small number of clinical trial participants, yes — and it is life-changing. Implanted interfaces let people with paralysis move cursors, type and control devices by intent. For everyone else, non-invasive consumer headsets remain crude: a few commands, lots of calibration.

Not as a normal service. Electric air taxis have flown crewed test flights and secured early certification steps, with launch routes announced in a few cities. Commercial passenger operations at meaningful scale have not started.

No. Single-purpose home robots — vacuums, mowers, window cleaners — are genuinely good. A general machine that picks up whatever your family left on the floor, decides where it belongs and puts it away is a research problem, not a product.

No. Humanoids are advancing quickly in warehouses and pilot deployments, but general household work is unsolved. Early units that exist are expensive, supervised, narrow in skill and mostly sold to businesses or developers.

No. Researchers print tissue patches, cartilage, skin grafts and organoids used for drug testing, and a few simple structures have been implanted in trials. A printed, vascularised, transplantable kidney or heart is not close.

No. Current machines are far too small and noisy to factor real keys. The genuine near-term risk is 'harvest now, decrypt later', which is why post-quantum cryptography migration is already underway in serious organisations.

No. Experiments have achieved scientific energy gain in the reaction itself, which is a genuine milestone, but no facility produces net electricity for a grid. Commercial pilot plants are announced for the 2030s and those timelines have slipped before.