Posted by: Team Outlander
Posted on 06/23/2021
On April 23, 2005, ‘Me at the Zoo’ was the first video ever uploaded to a then little-known site called Youtube. This moment was an inflection point that precipitated the explosions of video-first social media services over the next 15 years: TikTok, Instagram, Snapchat, etc. Moreover, it was a harbinger of much broader changes to come.
COVID-19 altered the world in many ways; being locked in our homes forced us to rethink how we communicate, work, and socialize. It has also led to a tectonic shift in how businesses operate. COVID has been the accelerant that has led to the decentralization of the global workforce.
This is not a discrete or temporary event. As the pandemic recedes and the world begins to regain an air of normalcy, many businesses have been compelled by their employees to rethink the very concept of ‘the office.’ The ideas put forth by managers for generations—such as productivity tied to the physical place of work—have now been debunked as relics of the industrial age. This shift did not transpire out of some charitable concern for employee welfare but, instead, driven by sheer economics and by employees refusing to work in conditions that negatively affect their lives without clear benefits to the productivity of their employers.
Robotic process automation is the early precursor of this ethos. With the automation of basic digital workflows and no-code environments, the automated back-office tasks increase productivity and free employees of time-consuming, remedial tasks. Companies such as UI-Path, Automation Anywhere, and others have trailblazed no-code automation tools that enable workers to automate their daily tasks. Much like the wheel or the steam engine, these tools shift the burden of work off of people onto machines. However, while videos have become ever more present in work, leveraging the information trapped within them has not been possible until recently.
The rapid development of machine vision and similar technologies propelled by the expansion of deep learning systems has primed the world for automations that use video as a proxy for many tasks that otherwise would be impossible to capture.
From sales information in Zoom presentations to quality control on semiconductor manufacturing lines, video data is ripe with hooks that can be used as proxies for automation.
Utilizing video data has never been more relevant. COVID-19 not only reframed the way we work but also the way we consume information on a day-to-day basis. Being quarantined in our homes forced us to consume our news, connect with the world, and seek entertainment not through the lens of personal interaction but primarily through videos broadcasted live and published on social media.
This drastic change in information consumption has led to a media landscape ripe for “disinformation” to propagate and pass into our society completely unchecked. Through a combination of state-sponsored misinformation and individual bad actors, it’s become incredibly difficult to understand what is true and what is not, and video is the most compelling piece of many disinformation arcs. It’s becoming increasingly easy for a maligned actor to edit a piece of video, remove valuable context, or share a video post from an entirely separate location with a leading caption and have their post go viral.
To address this type of disinformation, the ability to attribute provenance at scale is crucial. Consequently, the ability to collect live, social, and archival content is necessary for anyone attempting to validate whether or not a piece of content is original, manipulated, or has appeared across other channels. Finally, there is a need for automated tools to analyze and uncover strategies, patterns, and methods used by these actors in their narratives.
Large platforms such as Facebook, Google, and Microsoft have already begun developing in-house tools for tackling things like DeepFakes. Smaller organizations, however, are lagging behind in this effort, as they lack the expertise required to develop such systems.
For example, we developed a solution for the Association for Securing Democracy’s Hamilton 2.0 Dashboard to analyze large amounts of media content—both text and video—to comprehensively look at the information landscape created by state-sponsored media outlets. These types of technologies are the first step in a long journey towards a robust understanding of the current media landscape.
These are the problems that we at Vidrovr are focused on solving. We build AI tools for unlocking insights trapped in unstructured multimedia and video to drive data visibility, understand the video information landscape, and enable business automations. Learn more about our solutions at https://vidrovr.com.
We’re always looking for partners to help on this journey, so please reach out to our team at firstname.lastname@example.org. Plus, we’re hiring for our marketing, sales, and engineering teams! If you think you might be a good fit or want to learn more about our vision for the future, we’d love to hear from you.
As we explore the unknown of each new investment, our Field Guides are where we document all that we learn along the way.
So, whether you’re actively raising, trying to break into VC, or interested in our game-changing portfolio, our Field Guide's got you covered.
Sign up now for exclusive access to funding opportunities, events/resources from our network of experts, updates from our portfolio, and more!