The rate of digital content creation has exploded and artificial intelligence promises automation, personalization, and scale. Walter Frick at Harvard Business Review explained why AI can’t write articles yet in 2017, but also concluded that it can help us better write our own.
The following list features existing use cases where AI is leveraged to automate, assist or augment any step of the content creation process:
To deal with the growing volume of information and gain competitive advantage, the news industry has started to explore and invest in news automation.
- Washington Post – “Heliograf is creating a new model for hyperlocal coverage,” said Jeremy Gilbert, The Post’s director of strategic initiatives. “In the past, it would not have been possible for The Post to staff more than a handful of the most significant games each week. Now, we’ll be able to cover any game that we have data for, giving the teams and fans near-instant coverage to read and share.”
- Associated Press – The Associated Press uses Wordsmith to transform raw earnings data into thousands of publishable stories, covering hundreds more quarterly earnings stories than previous manual efforts.
- Reuters – Reuters Tracer automates end-to-end news production using Twitter data. It is capable of detecting, classifying, annotating, and disseminating news in real time for Reuters journalists without manual intervention.
Advertising copy generation
As ad-tech companies develop a better understanding of audience preferences, they will aim to personalize advertising messaging as much as possible.
- Saatchi Trained IBM Watson to Write Thousands of Ads for Toyota: Saatchi wrote 50 scripts based on location, behavioral insights and occupation data that explained the car’s features to set up a structure for the campaign. The scripts were then used to train Watson so it could whip up thousands of pieces of copy that sounded like they were written by humans.
- Persado enables brands to engage each consumer across all channels with personalized emotional language. “Imagine having a data scientist and a copywriter for each person in your audience; you get the language that performs and the analytics explaining why, resulting in more business and unseen insights”.
Think telling a story out of an Excel spreadsheet.
- Narrative Science creates Intelligent Narratives, which are human readable stories generated from quantitative data.
- Automated Insights’s website claims that hundreds of companies generate more than 1.5 billion pieces of content per year with their Wordsmith tool. Here is a gallery of examples.
Understanding and Producing Video
Video is among the fastest growing forms of content creation. Extracting meaning from videos and automatically creating new ones will be a growing opportunity.
- Reely – computer vision and deep learning specialized in transforming video into actionable data.
- Wibbiz – automatically creates premium videos from text.
Robo-writing & Paraphrasing
Input a topic and an AI will craft a human-readable article for you. This is what a few startups are building to help writers scale their content production. Examples include Articoolo, AI Writer and WordAI.
SEO and keyword research
Search engines now analyze your website through the lense of AI algorithms, which means they understand the overall context and theme, not just keywords. New AI tools for keyword research help you expand the scope of your keyword research process.
- Marketmuse – MarketMuse is an AI-powered research assistant that accelerates content creation and optimization so you can win in organic search.
- Twinword – an AI-powered keyword research tool that uses smart filters to quickly narrow down your keywords.
When you are doing research, it is easy to waste time browsing sources that are not relevant to you. Summarization helps you quickly skim articles and decide whether they are worth reading.
- Agolo – Agolo uses artificial intelligence to create summaries from your information in real-time.
- Salesforce’s has published ground-breaking research related to summarizing long-form text.
What if AI could do all the image editing for us?
- Object AI’s technology automatically detects the subject of the photo and makes photo editing intelligent.
- Adobe’s prototype AI tools let you instantly edit photos and videos
One big area of content creation includes email. Can AI help us create better content personalized to our audience?
- Crystal Knows – it understands the personality of your co-workers by analyzing their email language. It then helps you optimize your emails accordingly.
Voice recognition for note-taking
A lot of knowledge may get forgotten in conversations and meetings. Can AI help us turn voice into comprehensive text notes?
- Tetra – Tetra uses AI to take notes on phone calls, to help you focus, remember the details, and keep your team in sync.
Proofreading and grammar
AI models trained to make us better writers.
- Atomic Reach’s editor provides real time, custom and predictive recommendations to create perfectly written content with your audience in mind every time.
- Grammarly – Grammarly makes sure everything you type is clear, effective, and mistake-free.
Bots like Siri or Google Assistant are capable of answering fact-based questions, like “What is the capital of Spain?”. However, they often struggle to reply open-ended questions, like “Why is unemployment growing in Spain?”. In this situation, an AI would have to use multiple small snippets of evidence to make a combined judgement.
- IBM proposes to navigate through knowledge sources like Wikipedia to search for the answer, prioritizing some documents in their search over others; and combining knowledge from different parts of the documents they read to reason out an answer.
Can computers ask intelligent questions?
- Maluuba (Microsoft) – while there have been many advances in machine reading comprehension to develop models that can answer questions, Maluuba has been working to teach machines to ask informative questions.