Otto365 AI Brief #021

Amazon: Bedrock turns your videos into searchable moments

Describe the moment, Bedrock finds it in video

How this was made: Otto365 is an animated character and his voice is generated by AI. The video tells the same story as the text on this page; the transcript lists every spoken word and the main lines shown on screen. Spotted an error?
  • Published
  • By the Otto365 AI Brief. Checked and approved by Peter Ferguson, Editor, on . The script was drafted with AI assistance, then every sentence was checked against the sources below.
  • Announced 19 Oct 2026
  • Voiced by AI

What happened

Amazon's cloud can search hours of video for the moments you describe. AWS announced TwelveLabs Pegasus 1.5 on Amazon Bedrock on the 9th of October 2026, with a developer guide on the 19th of October 2026. It turns long video into timestamped data about the moments you choose.1

Three things to know

  1. Segments with times First, you describe the moments you want, like speaker changes or brand appearances, and Pegasus returns every matching segment with start and end times. No pre-indexing: videos go straight from Amazon S3.1
  2. Search by image Second, a prompt can include a reference image: show it a product, logo or person, and it returns only the segments where that item appears. One request can carry up to 30 images.1
  3. Developer tool, Oregon Third, this is a developer tool, called through the Bedrock API, taking videos up to two hours long. AWS offers it only in its Oregon Region in the United States, billed through AWS Marketplace.1

Why it matters to you

Otto365's take (opinion)

So, my take. Your existing video, webinars, demos, event footage, is becoming searchable by describing what you want. You probably won't call the API yourself, but expect it in your video tools.1

Do this week

Try this: write down three moments you would want to find in your videos.

Sources

  1. AWS: Pegasus 1.5 from TwelveLabs is now available in Amazon Bedrock | Artificial Intelligence (opens in a new tab) Primary source, accessed 20 Oct 2026
The quotes we checked (25)

  • On Oct 9, 2026 we announced the availability of TwelveLabs’ Pegasus 1.5 on Amazon Bedrock.

    AWS · source 1

  • on 19 OCT 2026 in Amazon Bedrock, Announcements, Foundational (100)

    AWS · source 1

  • Pegasus 1.5 is a video language model that turns long-form video into structured, timestamped metadata.

    AWS · source 1

  • You describe the moments that matter to your workflow, and the model returns every matching segment as JSON.

    AWS · source 1

  • yet most of it remains unsearchable and underutilized

    AWS · source 1

  • Now, Pegasus 1.5 analyzes the entire video, finds every segment that matches user-defined criteria, and returns structured, timestamped metadata.

    AWS · source 1

  • With earlier versions of Pegasus, you can point the model at a clip, ask a question, and get an answer.

    AWS · source 1

  • such as editorial segments, sports plays, speaker changes, or brand appearances. Pegasus 1.5 returns every matching segment as JSON, with start and end times and the requested fields populated.

    AWS · source 1

  • This makes it possible to locate a specific product, logo, or person, along with what they are doing on screen.

    AWS · source 1

  • Pegasus 1.5 introduces multimodal prompting, so you can include reference images alongside your text prompts.

    AWS · source 1

  • A single request can carry up to 30 reference images.

    AWS · source 1

  • In segmentation mode, the reference image lives inside the definition, so Pegasus returns only the segments where that product appears.

    AWS · source 1

  • Does the object in the reference image appear in this video?

    AWS · source 1

  • Yes, the green canoe appears in the video.

    AWS · source 1

  • No ingestion pipeline or pre-indexing step is required. Videos are sent directly from Amazon S3 and results are returned in the response.

    AWS · source 1

  • Pegasus 1.5 analyzes videos up to two hours long in a single request, up from one hour in Pegasus 1.2.

    AWS · source 1

  • You call it on the bedrock-runtime endpoint with the InvokeModel API, or with InvokeModelWithResponseStream when you want the response streamed as it’s generated.

    AWS · source 1

  • TwelveLabs Pegasus 1.5 is available today on Amazon Bedrock in the US West (Oregon) AWS Region.

    AWS · source 1

  • Pegasus 1.5 is billed through AWS Marketplace, so its charges appear on your AWS bill and in AWS Cost Explorer under TwelveLabs rather than Amazon Bedrock.

    AWS · source 1

  • The following example identifies each editorial section in a news broadcast and returns a title, summary, and key topics for each one:

    AWS · source 1

  • Pegasus 1.5 is designed for organizations that manage large volumes of long-form video:

    AWS · source 1

  • Their work shifts from finding and tagging every moment by hand to reviewing and refining the metadata generated by the model.

    AWS · source 1

  • Extracting structured, actionable insights from that content has historically required labor-intensive manual review or fragmented toolchains.

    AWS · source 1

  • then catalog years of footage that was previously searchable only by transcript keywords.

    AWS · source 1

  • Instead of a prompt, you send one or more segment definitions, and the model returns every segment of the video that matches each one.

    AWS · source 1

Transcript

Full transcript

Amazon's cloud can search hours of video for the moments you describe.

Play from 0:04 AWS announced TwelveLabs Pegasus 1.5 on Amazon Bedrock on the 9th of October 2026, with a developer guide on the 19th of October 2026. It turns long video into timestamped data about the moments you choose.

Play from 0:21 First, you describe the moments you want, like speaker changes or brand appearances, and Pegasus returns every matching segment with start and end times. No pre-indexing: videos go straight from Amazon S3.

Play from 0:36 Second, a prompt can include a reference image: show it a product, logo or person, and it returns only the segments where that item appears. One request can carry up to 30 images.

Play from 0:50 Third, this is a developer tool, called through the Bedrock API, taking videos up to two hours long. AWS offers it only in its Oregon Region in the United States, billed through AWS Marketplace.

Play from 1:04 So, my take. Your existing video, webinars, demos, event footage, is becoming searchable by describing what you want. You probably won't call the API yourself, but expect it in your video tools.

Play from 1:19 Try this: write down three moments you would want to find in your videos.

I'm Otto365.

Follow the Brief.

On screen

Play from 0:00
Otto365 AI Brief #021, Amazon, 19 Oct 2026
Describe the moment, Bedrock finds it in video
Find moments in video
Play from 0:04
Pegasus 1.5 on Bedrock
Guide posted 19 Oct 2026
Long video
Describe moments
Pegasus 1.5
Timed segments
Source: AWS, 19 Oct 2026
Play from 0:21
Segments with times
Video from S3
Segment definitions
Pegasus 1.5
Times and fields
What comes back
Segment title
Summary
Key topics
Start and end
Play from 0:36
Search by image
Reference image
Text prompt
Pegasus 1.5
Where it appears
Does the object in the reference image appear in this video?
Yes, the green canoe appears in the video.
Play from 0:50
Developer tool, Oregon
Your code
Bedrock API
Pegasus 1.5
JSON reply
US West (Oregon) only
Billed via AWS Marketplace
Up to two hours per request
Play from 1:04
Searchable video
Before: Manual review, Keyword search
With Pegasus: Describe moments, Review results
Webinars and demos
Event footage
Describe what you want
Play from 1:19
List three moments to find
Voiced by AI. Checked by humans. Apps 365 sells AI and Microsoft 365 services. Not endorsed by companies named.

Voiced by AI. Checked by humans. Apps 365 sells AI and Microsoft 365 services. Not endorsed by companies named. Report an error.