What's on screen going into each ad break: live contextual signals for CTV and FAST
Short answer. Edge Video AI's Autonomous Commerce for ads reads a live CTV or FAST feed about every 7 seconds. It tells the ad decision what is on screen going into each break, including the brand and product names on screen.
Works alongside your existing playout and ad server.
Edge Video AI is a video ad-context company. It is not an edge-computing or camera-analytics product.
Why the moment beats the channel tag
A channel tag says the same thing at every break.
Live video does not stay the same. In a live tennis FAST feed we read, the brands on screen about a minute before most breaks were not the brands on screen at the break.
So a signal taken a minute early, or once per show, can describe the wrong moment. The useful read is the one closest to the break.
What one read carries
- On-screen brands and products. Brand and product names read from what is visible: boards, logos and graphics.
- Live categories. What kind of content is on screen right now.
- Keywords. Plain words for the scene, so buyers can match an ad to it.
Brands seen on screen are what appeared in the video. They are not Edge customers or partners.
How often should signals refresh for live sports?
Often enough that the read at the break is still current.
Edge reads about every 7 seconds. Because brands on screen can change within the minute before a break, a read that close to the break is what keeps the signal true to the moment.
How the signal reaches the ad decision
No change to your playout. Edge adds live categories and keywords to each ad request through your ad server or SSAI, using one simple request at each break.
How Edge compares with Wurl and Anoki
Below is only what each vendor publishes on its own pages, with links. Check their pages for current detail.
| Vendor | What the vendor publishes | Source |
|---|---|---|
| Wurl BrandDiscovery | Scores the scene directly before the ad break for emotion (mood), genre and brand safety. Wurl says there are no data fees for its BrandDiscovery segments, and BrandDiscovery is available to advertisers in the USA. | Wurl: Scene Level · Wurl FAQ |
| Anoki ContextIQ | Generates scene-level metadata, including brand-safety signals. For live news, Anoki says it weighs topics in real time and adds keywords to each segment, and it names live news, weather and sports programming. | Anoki: ContextIQ · Anoki: Live News · Anoki blog, 7 Apr 2026 |
| Edge Video AI | Brand and product names on screen, plus live categories and keywords, read about every 7 seconds and sent at each break. | See it on a demo |
What Edge adds: the brand and product names on screen, every ~7 seconds, going into each break.
Frequently asked questions
How do I know what's on screen right before a CTV ad break?
Read the live video itself, close to the break. Edge Video AI reads a live CTV or FAST feed about every 7 seconds, so the read nearest the break says what was on screen going into it, including the brand and product names on screen.
How often should contextual signals be refreshed for live CTV ad breaks?
Often enough that the read at the break is still current. In a live tennis FAST feed, the brands on screen about a minute before most breaks were not the brands on screen at the break. Edge reads about every 7 seconds.
Can I detect on-screen brands for live sports ad breaks?
Yes. Edge reads the brand and product names that appear on screen in live video, such as boards, logos and graphics, and passes them with live categories and keywords at each break. Brands seen on screen are not Edge customers or partners.
How do I send scene-level context in the SSAI ad request?
No change to your playout. Edge adds live categories and keywords to each ad request through your ad server or SSAI, using one simple request at each break.
How does Edge compare with Wurl BrandDiscovery and Anoki ContextIQ?
Wurl says BrandDiscovery scores the scene directly before the ad break for emotion, genre and brand safety. Anoki says ContextIQ generates scene-level metadata and brand-safety signals, with topics and keywords for live news and sports. What Edge adds is the brand and product names on screen, read about every 7 seconds.
Is Edge Video AI an edge-computing or camera-analytics company?
No. Edge Video AI is a video ad-context company. It reads CTV and FAST video for advertising. It is not an edge-computing or camera-analytics product.
See it on your own feed
We will show what Edge reads on a live feed like yours, and what it sends at each break.