Writing is One of The Most Vital Contributions of The Human Race
Starting from documenting our history, to passing on our teachings and information, our intellectual society has been found on the basis of human-written data. Our society has evolved on the foundation of millions of written scriptures.
The same humans, with brilliance at par, have now created algorithms, which can create content, without human intervention. The question that comes to mind, is whether machine generated writings will have the same essence of human writing.Will it be biased or honest? But the real question is, are we ready for it?
Yes, we are indeed, at least in some cases.
Take the example of business operations and intelligence, of any industry. Business reporting requires documenting down thousands of data and analysis. Now, we have BI tools with all sorts of dashboard and visualization features, but still the data popping up, needs to be given a language to be understood and analyzed. Natural Language Generator (NLG) swoops in to fill up this vital gap. It can digest large volumes of data in question and generate reports automatically, thus augmenting the job of business users so they can focus more on high-value tasks and less on menial work.
- Healthcare companies can use it to document patient data on a real time basis, by analyzing data sent through connected patient monitoring devices.
- Financial institutions can use it to flash real time stock updates and generate analytical reports.
- Educational organizations can use it to create assignments for students.
- Every marketing department can have a NLG tool, to create quick short contents for digital as well as physical media.
- Customer representation is already using this tool to make bots communicate with customers and is seeing an exponential rise in every sector as days go by.
Some popular tools are Narrative Science’s Quill, Arria’s Recount and Yseop’s Savvy. Most of these tools enable sales representatives with information on consumers so that they can sell the right product to the respective customer. Some tools have an inbuilt learning system featured. Companies who use them can train their NLG tools about a certain style or format of writing by feeding it examples.
But how about AI as an author or a journalist?
Do we dare to let them write an entire news article? How does it work?
The Post has used Heliograph, an AI that has to be just connected with a structured data source, and a template with keywords on which the article has to be based. It then mines the relevant data, matches it with the available keywords, maps and arranges the keywords in an order that creates understandable messages. It compiles and structures phrases, paragraphs or articles for publishing in various platform-relevant formats. The question of accountability of data or information can be nullified, as AI tools can verify and authenticate data in numerous ways. It’s most useful for journalists and writers in their secondary research about any topic. They can shake off the headache of data mining, data validation and analysis, which can be easily drawn out by AI. AI can be also used to dig and publish local and niche news aimed towards personalized journalism.
Does this mean it will replace journalists and writers?
Though AIs can give a complete unbiased and honest picture of an event or news, but we will still need the human touch to display the bigger picture derived from current state of events. AI can help a writer with a popular plot line, or a unique story line idea by mining through all the stories ever written online in past or at present, but we still need human writers to give certain philosophies to characters, insightful dialogues and natural elements to draw interpretations and reflections that races, ignites or lulls the mind and heart of readers.
Now when AI is generating film scripts, and many companies are investing in AI tools for content generation than recruiting new talents, human writing will become history, in a sense.