The future of office productivity
From saving time to breaking down barriers, the endless possibilities of AI technology
“This technology makes it possible to see a real-time transcript of any call in your native language, irrespective of whose at the other end.”
For decades, many Artificial Intelligence (AI) experts felt like a lone voice crying out in the wilderness, heralding the coming of something significant. Over the years, there have been many false starts and setbacks in the development of AI. But in recent years, we have seen tremendous advances in the field. But now, the mass adoption of AI technologies is finally here. These technologies are slowly changing every aspect of our lives, from education and healthcare to transportation and manufacturing. But unless you’ve been living under a stone, the tool that captured many’s imagination is undoubtedly chatGPT, the AIpowered chatBot capable of creating human-like outputs. But chatGPT is already passe since its successor, GPT-4, is already in use.
However, people do not realise the profound effects this will have on their productivity at work. So let me try to explain it.
When searching for information, our first port of call is undoubtedly Google, a traditional search engine. So let’s say someone wants to find the tallest mountain in the world. He would typically think about search terms such as the “tallest mountain”, type the search query, and the search engine will look through all the websites on the internet. It will then find the page that mentions the relevant words and rank the results based on their relevance to the search. The user must review those links, read through the pages and identify the needed information. In technical jargon, the process is called information retrieval.
On the other hand, Bing integrated GPT-4 with its search engine. The process is substantially different. The person would still think about search terms and type the search query, but the similarities stop there. Bing would immediately return the name of the tallest mountain. The process is called information extraction, which is much faster than what we’re traditionally used to.
But this is just the start. Just think about Office 365. Microsoft is integrating GPT-4 with it as a new feature called Copilot. This service will help users save time and effort in various ways while boosting productivity. Here are some concrete applications: • Writing assistance: Copilot can assist in the writing process by automatically adjusting spacing to improve legibility, adding speaker cues to help during presentations, and suggesting improvements to strengthen the proposal. For example, when creating a customer proposal, Copilot can quickly use customer notes and internal documents to make a first draft, insert product images, and pull relevant visuals from other files. It can even generate an FAQ section for the proposal. Furthermore, since Copilot can also suggest ideas, writers can finally bid farewell to the frustration of writer’s block.
• Presentation creation: Copilot can quickly turn a Word document into a PowerPoint presentation, complete with professional design treatments and speaker notes. High-resolution original visuals can be generated on the fly using powerful AI models. Users can add slides and give Copilot instructions to make the presentation more visual or animated. This feature saves time and effort in creating engaging presentations for meetings and pitches.
• Data analysis: Copilot can analyse data in Excel and generate insights and actions. For example, if a user needs to analyse quarterly sales results, they can ask Copilot to analyse the data and give three key trends. It can then create a new sheet to drill deeper into one of the trends and visualise what contributed to the decline in sales growth. Furthermore, the application can even create a simple scenario model and generate a projected model graph.
• Email management: Copilot can help triage the inbox and prioritise important emails, saving users hours. It can also summarise long email threads and generate replies using data from an Excel file. Users can edit and make the response more concise, change the writing style, or add more context. This feature makes users’ lives easier by responding to important emails quickly and efficiently.
Overall, Copilot can increase productivity by automating repetitive tasks, suggesting improvements to strengthen work output, and enabling users to work more efficiently and effectively.
But you should keep in mind that this is just the start!
We know that when Microsoft bought Skype, the video-calling platform, they actively worked on real-time automated transcription. This technology makes it possible to see a real-time transcript of any call in your native language, irrespective of whose at the other end. Microsoft has another tool named VALL-E that can replicate a person’s voice from just three seconds of audio. So by combining the two technologies, one can easily listen to the other person speak in your native language, mediated by AI, even though he might not know a single word in your language. The era of Star Trek’s universal translator is around the corner.
Another impressive technology is AI-generated images which let the AI create any scene imaginably. However, users need help to make visual sceneries exactly how they imagine them. They must navigate through complex prompts, squeezing their neurons to devise ways of directing the algorithm. But work is ongoing to make it even easier by combining GPT-4 with image models like DALL-E. Some even allow users to remix different images as inputs, possibly rubbing off parts of an image and letting the system dream additional variants. Such technology is the first step, and at the moment, there is already active research to create AI-generated movies.
As we look at the future, it is becoming clear that the possibilities unleashed by AI and its related technologies are limitless. From saving us time and effort in our daily tasks to breaking down language barriers, the potential for these technologies to change our lives is immense. With the introduction of tools like Copilot, we’re already seeing the transformative effects of AI in the workplace, and it’s only a matter of time before we see similar advancements in other fields. As we continue to push the boundaries, one thing is for sure: the future of AI is both exciting and unpredictable, so expect more disruptions in the coming months.