1001 Smart Cities
🇬🇧 London

Vivacity AI Traffic-Sensor Network

🤖 AI-generated summary · ◆ 13 sourced claims · 🔎 Last verified

This overview was generated with AI from the public sources listed below and checked against them before publication. It has not had a complete human fact-check — treat it as a starting point and follow the sources.

13 sourced claims on this page, none individually verified yet — turn on “Sources in the text” to see each one.

How this atlas is made · What “last verified” means · What “source-checked” means · The full record behind this page

TfL has counted road users with edge-AI cameras since 2018 — video classified on the device, then discarded. But the headline "over 1,000 cameras" is a borough-owned pool TfL is still negotiating access to, not its own estate.

Outcome🔄 Ongoing
Evidence🏛️ Official
what does this mean?The strongest source is the operator's or the city's own reporting — an official report, dataset or press release. It grades who did the looking — not how well the project went. All four grades →
Run byTransport for London (sensors supplied by Vivacity Labs)
Scale🏙️ City-wide

Part of 1001 Smart Cities — 1,167 documented projects · 🏙️ 15 more in London · 📡 Sensors & IoT guide · 🚲 Mobility projects

Transport for London has used Vivacity Labs' edge-AI cameras since 2018, first at two Millbank sites, later at 43 sensors across 20 central London locations. They classify pedestrians, cyclists, wheelchair users and vehicles on-device and discard the video; TfL's competitive trial found them up to 98% accurate. TfL's January 2026 strategy expands the estate and seeks a borough data-sharing agreement covering over 1,000 cameras. The firm states it has never used facial recognition.

Live 📅 since 2018 🤖 Uses AI

📊 Impact

TfL's competitive trial at two Millbank locations found the sensors up to 98% accurate at classifying road users; TfL then introduced 43 more sensors at 20 central London locations (TfL, January 2020).

🎓 Lesson

A vendor's installed base is not a city programme. The 1,000-odd Vivacity sensors in London are split between TfL and twenty-five London councils, each procuring its own — which is why TfL had to negotiate a data-sharing agreement to see them at all.

🚲 MobilitySensors & IoTArtificial intelligenceTraffic managementCyclingPrivacy & personal data

Sources

↗ TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)Official primary. Carries every load-bearing fact: 'Since 2018, TfL has trialled using Vivacity Labs sensors at two busy locations along Millbank'; 'up to 98 per cent accurate'; 43 more sensors at 20 central London locations; 'All video captured by the sensors is processed and discarded within seconds'; AI mode classification.
↗ TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)Official primary for the expansion. Exact wording: 'TfL is working closely with boroughs on a data-sharing agreement, which will give access to insights from over 1,000 cameras.' Future tense, borough-owned pool. FUSION is a separate item in the same release and is NOT stated to take Vivacity data.
↗ Vivacity Labs — Data Privacy & SecurityVendor statement on the privacy design: video processed locally and discarded, no personal data, 'never done any work involving facial recognition'. Also discloses the exceptions: up to an hour of video stored during setup, and a 10-minute clip per sensor retained for the deployment's lifetime to test software updates.
↗ Smart Cities World — VivaCity partners with 25th London councilPublished 21 November 2022Trade press evidencing the borough-by-borough footprint: Bexley is the 25th London council to contract Vivacity, installing 42 sensors in place of induction loops.
↗ Vivacity Labs — Borough Safer Streets & Better Bus Partnerships (vendor page)Published 23 July 2024Cited as the ORIGIN of two figures that must not be read as independent: 'approximately 1,000 live VivaCity sensors across London supporting TfL and London boroughs' and 'Independently verified by Transport for London, VivaCity sensors achieve a market-leading 97% accuracy rate.' TfL's own release says 'up to 98 per cent' — the 97% is vendor copy.

📅 How old is the evidence?

Two clocks, kept apart: when this entry’s sources were published, and what has happened to its record in the atlas. Documented span: since 2018. 4 of 5 sources carry a publication date, from 16 January 2020 to 26 January 2026.

When the sources were published

  1. Transport for London · operatorTfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
  2. Smart Cities World · pressSmart Cities World — VivaCity partners with 25th London council
  3. Vivacity Labs · vendorVivacity Labs — Borough Safer Streets & Better Bus Partnerships (vendor page)
  4. Transport for London · operatorTfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)

Every source, with links ↑

What happened to this record

  1. Sources last read
  2. Entered the record

Every revision, permanent →

Show your work What this page rests on and what it can’t tell you · all 13 claims with their sources · the forensic record

Evidence Passport · public beta 13 addressable claims, 5 source dossiers, one permanent baseline. The passport is unreviewed — an AI-assisted structural migration, not a human or expert endorsement. Every revision keeps a permanent copy under revision history.

Inspect the claims, one by one — verbatim text, sources and caveats for all 13

The text below is copied verbatim from the published record. Each claim names the source dossiers that carry it; caveats stay attached to the claim they qualify.

Claim 1 clm-306337ffdc95917d quantitative
Transport for London has used Vivacity Labs' edge-AI cameras since 2018, first at two Millbank sites
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
TfL's own release states: 'Since 2018, TfL has trialled using Vivacity Labs sensors at two busy locations along Millbank.' The 'edge' description comes from the supplier's privacy page, which says the system 'processes all video locally, produces anonymous data feeds and discards the video within milliseconds'.
Caveat
TfL describes the two Millbank installations as a trial rather than settled operational use, and neither source confirms that those two sites are still running today.

Sources carrying this claim

  1. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
  2. Vivacity Labs — Data Privacy & Security
Claim 2 clm-85e80fb7df86344b quantitative
later at 43 sensors across 20 central London locations
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The January 2020 release says TfL 'is in the process of introducing 43 more Vivacity sensors at 20 central London locations to gather data and further test', and lists all twenty sites in its notes to editors under the heading 'The proposed 20 central London locations for the 43 Vivacity sensors are:'.
Caveat
At the time of writing TfL called the rollout a work in progress and the twenty sites 'proposed'. No listed source confirms that all 43 sensors were installed, and the release counts them as 43 sensors in addition to the two at Millbank, not as the whole estate.

Sources carrying this claim

  1. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
Claim 3 clm-99f84ae395caeb9c descriptive
They classify pedestrians, cyclists, wheelchair users and vehicles on-device
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
TfL's 2026 strategy release lists the classes: 'Vivacity technology uses AI to distinguish between different modes of travel including walking, cycling, wheelchair use, taxis, and heavy goods vehicles.' The 2020 release names the vehicle classes as 'cars, HGVs, vans, motorcyclists and buses' alongside people cycling and walking. The supplier's privacy page describes where the work happens: the system 'processes all video locally'.
Caveat
Wheelchair users appear only in the 2026 release; the 2020 trial release does not list them. That the classification runs on the device rather than in the cloud is stated by the supplier, not by TfL.

Sources carrying this claim

  1. TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)
  2. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
  3. Vivacity Labs — Data Privacy & Security
Claim 4 clm-15252b96ec19eede descriptive
discard the video
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
TfL states: 'All video captured by the sensors is processed and discarded within seconds, meaning that no personal data is ever stored.' The supplier's privacy page says the system 'processes all video locally, produces anonymous data feeds and discards the video within milliseconds'.
Caveat
The supplier's own page names two exceptions to this: during installation it may store up to an hour of video to calibrate the sensor, and a ten-minute clip from each sensor is kept for the lifetime of the deployment so that software updates can be tested.

Sources carrying this claim

  1. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
  2. Vivacity Labs — Data Privacy & Security
Claim 5 clm-237012618dd49981 quantitative
TfL's competitive trial found them up to 98% accurate
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
TfL reports the result of its own procurement trial: 'The competitive trial, which is the first time the technology has been used within London, showed that the Vivacity sensors are up to 98 per cent accurate.' The supplier's marketing page gives a different number for what it presents as the same TfL assessment: 'Independently verified by Transport for London, VivaCity sensors achieve a market-leading 97% accuracy rate.'
Caveat
This is TfL reporting on a trial it ran itself; no independent body has published an accuracy assessment. TfL does not say what was measured, over how long, or against which reference count, and 'up to' names a best case rather than a typical one.

Sources carrying this claim

  1. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
Claim 6 clm-2bac69c396df60dc quantitative
TfL's January 2026 strategy expands the estate
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The 'London on the move' release, dated 26 January 2026, lists among the measures TfL has committed to: 'TfL is expanding the use of Vivacity cameras across London to better understand how people and vehicles move through the city.'
Caveat
The strategy states the intention to expand but names no target number of sensors, no locations and no date by which the expansion is to be complete.

Sources carrying this claim

  1. TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)
Claim 7 clm-81dc04f6178246e0 quantitative
seeks a borough data-sharing agreement covering over 1,000 cameras
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The same release says: 'TfL is working closely with boroughs on a data-sharing agreement, which will give access to insights from over 1,000 cameras, highlighting the importance of collaboration in shaping London's future mobility.'
Caveat
TfL writes in the future tense: the agreement is being worked on and would give access to those insights. No listed source says it has been concluded, and the release does not state that all of those cameras are Vivacity sensors.

Sources carrying this claim

  1. TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)
Claim 8 clm-ace924ce2a6da957 descriptive
The firm states it has never used facial recognition.
Published field
snippet · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The supplier's privacy page says: 'We have also never done any work involving facial recognition and we will never do so as it violates our commitment to privacy.'
Caveat
This is the supplier's own statement about its own product. No listed source records an external audit or a regulator's finding that tested it.

Sources carrying this claim

  1. Vivacity Labs — Data Privacy & Security
Claim 9 clm-2c2a897e873c7b54 quantitative
TfL's competitive trial at two Millbank locations found the sensors up to 98% accurate at classifying road users
Published field
impact · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
TfL's January 2020 release ties the trial to the two Millbank sites and reports the figure: 'Since 2018, TfL has trialled using Vivacity Labs sensors at two busy locations along Millbank' and 'The competitive trial ... showed that the Vivacity sensors are up to 98 per cent accurate.' The supplier's own page attributes a 97 per cent figure to Transport for London instead.
Caveat
The figure is TfL's account of a trial it commissioned and ran; it has not been audited or published in full. TfL gives no measurement period, sample size or reference count, and 'up to' describes the best result rather than an average.

Sources carrying this claim

  1. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
Claim 10 clm-7a7d0cb5a71ab272 quantitative
TfL then introduced 43 more sensors at 20 central London locations (TfL, January 2020)
Published field
impact · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The release, dated Thursday 16 January 2020, says TfL 'is in the process of introducing 43 more Vivacity sensors at 20 central London locations to gather data and further test to understand the full range of capabilities the technology has to offer', and names all twenty proposed sites.
Caveat
The wording is 'in the process of introducing', and the notes to editors call the twenty sites 'proposed'. The entry reads this as completed; no listed source reports the finished installation.

Sources carrying this claim

  1. TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020)
Claim 11 clm-5134dbf45fa97ac3 quantitative
The 1,000-odd Vivacity sensors in London
Published field
lesson · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The figure originates with the supplier, whose page states: 'Currently, there are approximately 1,000 live VivaCity sensors across London supporting TfL and London boroughs with high quality transport data.' TfL's 2026 strategy separately refers to 'insights from over 1,000 cameras' covered by the planned borough data-sharing agreement.
Caveat
The count is the supplier's own, published in July 2024, and no independent tally exists. TfL's figure is a count of cameras reachable through a planned agreement, not a statement about how many Vivacity sensors are installed.

Sources carrying this claim

  1. Vivacity Labs — Borough Safer Streets & Better Bus Partnerships (vendor page)
  2. TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)
Claim 12 clm-83aec6076f33ffca descriptive
are split between TfL and twenty-five London councils, each procuring its own
Published field
lesson · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
The trade report is headlined 'VivaCity partners with 25th London council to tackle congestion' and its opening summary says: 'The company will deploy 42 AI sensors in the Bexley area that can differentiate between different modes of transport and report on the make-up of traffic.' Each council contracts the supplier separately. The supplier describes the London estate as 'supporting TfL and London boroughs'.
Caveat
The trade report establishes twenty-five London councils as of November 2022, not more than twenty-five, and the rest of that article sits behind a registration wall. No listed source breaks the London estate down by owner, so the share held by TfL rather than the boroughs is not documented.

Sources carrying this claim

  1. Smart Cities World — VivaCity partners with 25th London council
  2. Vivacity Labs — Borough Safer Streets & Better Bus Partnerships (vendor page)
Claim 13 clm-de18ede002d90f76 descriptive
which is why TfL had to negotiate a data-sharing agreement to see them at all
Published field
lesson · verbatim
Mapping
mapped · no interpretation or interpolation added by this passport
What the mapping says
TfL's 2026 strategy confirms the negotiation: 'TfL is working closely with boroughs on a data-sharing agreement, which will give access to insights from over 1,000 cameras, highlighting the importance of collaboration in shaping London's future mobility.'
Caveat
TfL presents the agreement as collaboration rather than as a consequence of fragmented ownership, and does not say that it is currently unable to see borough data. The causal reading is the atlas entry's, not TfL's.

Sources carrying this claim

  1. TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026)
Forensic record — source dossiers, gaps, hashes and the exact revision this page renders

This layer records what is known about each source, what is still missing, and the exact snapshot this page is rendering. A missing date, archive or hash stays visible as missing.

unreviewed Revision 2026-08-21-evidence-baseline effective 2026-08-06 · created 2026-08-22

Awaiting operator and domain-expert review; publication of this revision implies neither.

sha256:728d2ecbe922c02802f30beba1af9841e794ab1f0ebad252914a45602b9d57e1
Revision history → Permanent version link → Machine-readable record → Evidence Constitution →
13
of 13 claims mapped
5
source dossiers
4
publication dates recorded
0
archives linked
0
exact locators
0
content captures hashed

Governance facts carried by the record

Operator
Transport for London (sensors supplied by Vivacity Labs)
Actor types
city-government, private
Funding model
public
Funding description
Not recorded
Cost
Not recorded
Ownership
Not recorded
Decision rights
Not recorded
Exit conditions
Not recorded

5 source dossiers

TfL newsroom — Artificial intelligence to help fuel London's cycling boom (16 Jan 2020) tfl-newsroom.prgloo.com · 7 linked claims

Open original source ↗

Stable source ID
src-0a5c39329dc7647e
Publisher
Name not classified · tfl-newsroom.prgloo.com
Type / language
web-page · language not classified
Published
2020-01-16
Updated
Not recorded
Source last checked
No source-level check date
Record last verified
2026-08-06
Dossier created
2026-08-22
Retrieved copy
No local capture recorded
Archive
No archived copy linked
Content hash
No captured content hash
Locator
Open the recorded source note

Official primary. Carries every load-bearing fact: 'Since 2018, TfL has trialled using Vivacity Labs sensors at two busy locations along Millbank'; 'up to 98 per cent accurate'; 43 more sensors at 20 central London locations; 'All video captured by the sensors is processed and discarded within seconds'; AI mode classification.

Descriptive only — not page/paragraph exact
Claims linked
TfL newsroom — 'London on the move' five-year strategy (26 Jan 2026) tfl-newsroom.prgloo.com · 5 linked claims

Open original source ↗

Stable source ID
src-620ccd907f4f74aa
Publisher
Name not classified · tfl-newsroom.prgloo.com
Type / language
web-page · language not classified
Published
2026-01-26
Updated
Not recorded
Source last checked
No source-level check date
Record last verified
2026-08-06
Dossier created
2026-08-22
Retrieved copy
No local capture recorded
Archive
No archived copy linked
Content hash
No captured content hash
Locator
Open the recorded source note

Official primary for the expansion. Exact wording: 'TfL is working closely with boroughs on a data-sharing agreement, which will give access to insights from over 1,000 cameras.' Future tense, borough-owned pool. FUSION is a separate item in the same release and is NOT stated to take Vivacity data.

Descriptive only — not page/paragraph exact
Claims linked
Vivacity Labs — Data Privacy & Security vivacitylabs.com · 4 linked claims

Open original source ↗

Stable source ID
src-1cc378979008e482
Publisher
Name not classified · vivacitylabs.com
Type / language
web-page · language not classified
Published
Not recorded
Updated
Not recorded
Source last checked
No source-level check date
Record last verified
2026-08-06
Dossier created
2026-08-22
Retrieved copy
No local capture recorded
Archive
No archived copy linked
Content hash
No captured content hash
Locator
Open the recorded source note

Vendor statement on the privacy design: video processed locally and discarded, no personal data, 'never done any work involving facial recognition'. Also discloses the exceptions: up to an hour of video stored during setup, and a 10-minute clip per sensor retained for the deployment's lifetime to test software updates.

Descriptive only — not page/paragraph exact
Claims linked
Smart Cities World — VivaCity partners with 25th London council smartcitiesworld.net · 1 linked claim

Open original source ↗

Stable source ID
src-502da598fb72dd56
Publisher
Name not classified · smartcitiesworld.net
Type / language
web-page · language not classified
Published
2022-11-21
Updated
Not recorded
Source last checked
No source-level check date
Record last verified
2026-08-06
Dossier created
2026-08-22
Retrieved copy
No local capture recorded
Archive
No archived copy linked
Content hash
No captured content hash
Locator
Open the recorded source note

Trade press evidencing the borough-by-borough footprint: Bexley is the 25th London council to contract Vivacity, installing 42 sensors in place of induction loops.

Descriptive only — not page/paragraph exact
Claims linked
Vivacity Labs — Borough Safer Streets & Better Bus Partnerships (vendor page) vivacitylabs.com · 2 linked claims

Open original source ↗

Stable source ID
src-fa6fbeb1524c9c52
Publisher
Name not classified · vivacitylabs.com
Type / language
web-page · language not classified
Published
2024-07-23
Updated
Not recorded
Source last checked
No source-level check date
Record last verified
2026-08-06
Dossier created
2026-08-22
Retrieved copy
No local capture recorded
Archive
No archived copy linked
Content hash
No captured content hash
Locator
Open the recorded source note

Cited as the ORIGIN of two figures that must not be read as independent: 'approximately 1,000 live VivaCity sensors across London supporting TfL and London boroughs' and 'Independently verified by Transport for London, VivaCity sensors achieve a market-leading 97% accuracy rate.' TfL's own release says 'up to 98 per cent' — the 97% is vendor copy.

Descriptive only — not page/paragraph exact
Claims linked
Known limits of this passport
  • Publisher names and source languages have not yet been classified; domains are recorded without guessing.
  • Existing source notes are descriptive locators, not page- or paragraph-exact citations.
  • A missing archive, publication date or content hash is shown as null rather than inferred.
  • The project-level evidence grade ranks the strongest source in the entry; it is not a truth score for every claim.

Continue exploring

Compare this project

🇬🇧 London Adaptive Traffic Signals (FUSION / Real Time Optimiser) · London TfL's 2026 strategy expands both, but FUSION's AI-sensor integration is a future phase — Vivacity data does not yet retime signals. 🇻🇳 HCMC Intelligent Traffic Control Center · Ho Chi Minh City AI traffic cameras, opposite privacy models: Vivacity counts road users anonymously; HCMC's read plates to fine drivers (2020-). 🇬🇧 Emerging Technology Charter & London Privacy Register · London Vivacity's anonymous, no-facial-recognition cameras are exactly the public-space sensor London's Privacy Register was built to disclose.

Four cities count cyclists. Only one of them gives the count back.

Counting bicycles is the easy part, and every one of these does it well. The ladder runs from a number nobody acts on to a signal plan that hands the cyclist the green light.

ProjectSinceStatusOutcomeEvidenceRun byCostWhat happens to the count
Plan Vélo Digital Infrastructure MeshParis, 🇫🇷 2021 Live Ongoing Official City government not public Published. Permanent counters feed hourly totals into the open data portal on a rolling window, so anyone can audit whether the new lanes moved anything.
Vivacity AI Traffic-Sensor NetworkLondon, 🇬🇧 2018 Live Ongoing Official City government, Private company not public Classified on the device and the video thrown away — a deliberate privacy design, and a count split across two dozen boroughs that each bought their own.
Traffic Pilot — City-Wide Signal PredictionFrankfurt am Main, 🇩🇪 2022 Live Ongoing Official City government not public Handed back as advice: the app reads live signal phases and tells the rider what speed catches the green wave.
Green Wave for Cyclists (Grøn Bølge)Copenhagen, 🇩🇰 2007 Live Succeeded Official City government not public Built into the signal plan itself. The lights are timed to the cyclist rather than the cyclist to the lights — and the buses on those streets wait for it.

Explore the wider context

🚦 Asphalt or algorithms📖 Adaptive traffic signals📖 Automated speed enforcement (speed cameras)📖 Geofencing
📎 Cite this project

1001 Smart Cities (2026). “Vivacity AI Traffic-Sensor Network” — London, United Kingdom. The Smart City Atlas. https://1001smartcities.org/projects/london-vivacity-ai-sensors/ (last verified 2026-08-06). Data: CC BY 4.0.

The underlying data is free to reuse with attribution — see the open data page.

🚩 Report an error · 🔔 Follow changes to this entry (RSS)

✉️ Get the next source-checked smart-city project in your inbox

One short email a week: the Landmark, one Insight, the new cities. No spam, one-click unsubscribe. About the newsletter · RSS