TASQ AI Promises Faster Data Annotation for AI Development
TASQ AI is promising to speed up data annotation which is the boring but essential grunt work behind training machine learning models. If they deliver it could matter.
Data annotation is labeling images text and other content so AI systems can learn from it. Its tedious work often done by humans clicking through thousands of examples. TASQ wants to automate more of it.
TechCrunch covered the company and its approach to the annotation problem.
The AI industry has a dirty secret. Behind all the impressive demos are armies of workers doing manual labeling. Making that process faster and cheaper unlocks more AI development.
AI infrastructure companies dont get the headlines that consumer AI does but theyre just as important to the ecosystem.
TASQs approach uses existing AI to speed up annotation which is meta in a way. AI helping train better AI. The bootstrap problem in machine learning.
Competition in this space is fierce. Scale AI is the biggest player. Labelbox has funding. Amazon has its own solutions. TASQ needs differentiation to survive.
Whether faster annotation leads to better AI or just more AI is a question worth asking. Speed isnt always the right optimization.
Company profile 2021
