The future is being re-written: How machine intelligence will change the way we work
- SearchInk uses machine intelligence to extract, semantically connect and smart-export information contained in any type of document
- “What we are building here will have a substantial impact on the way we work and on our lives…”
- SearchInk is hosting a stand-out TOA satellite event on Friday 14th July: the Machine Intelligence Summit.
SearchInk is aiming to change your life. How? Well, if you’ve ever had to do a tedious data-entry task, you’ll have probably felt the instinctive inner response of “I wish someone else could do this.”
Brace yourself for the good news: soon, that someone else will be something else — because Sofie Quidenus, Founder and CEO of SearchInk, is creating a machine intelligence-based SaaS platform that will take care of it instead. Essentially, their machine intelligence-based product will read, sort, and act upon the type of documents and written forms that have — until now — required a human worker.
And it’s not just going to make life less repetitive on a human level. The bottom line is that this is the kind of technology that can make the sort of efficiency increases and cost reductions which most businesses can only dream of.
At TOA 2017 you can experience this future for yourself. As part of TOA’s Berlin-wide programme of Satellite events, Sofie and SearchInk are hosting a truly stand-out satellite event on Friday 14th July: the Machine Intelligence Summit.
In a day-long programme of events, some of the brightest minds in the world of machine intelligence will speak: it’s your ticket to understanding the future of data.
The afternoon session, from 13:00–18:00, is open to the public, and applications to attend are sure to fill up quickly — so make sure you apply now to grab your ticket for this fascinating event.
TOA’s Satellite Event programme comprises 200 events which scatter TOA’s unique approach to knowledge-sharing all over the city of Berlin.
And most of them are completely free, so it’s a great way to augment your TOA experience and learn something completely new — so browse the Satellite events schedule and buy your TOA ticket before the price increases on 30th June…
How do you explain what Searchink does, and why it’s important, to a newcomer?
Sofie Quidenus, SearchInk: SearchInk resolves a problem nearly every company has: how to deal with incoming requests and documents. To manage them and in order to fit them in internal systems and processes, a variable amount of manual work is required: manual data entry, manual corrections validation and reviews… it’s even more time and energy spent before the actual work could be done.
Using Machine Intelligence algorithms, SearchInk extracts and semantically connects the relevant data contained in highly diversified document streams. The result is a consolidated repository of information which can be subsequently used for further processing and Big Data analytics.
This concerns a lot of sectors: for example, Banking and Insurance, or General Accounting — where we offer semantic data extraction, clustering, structured and mapped export, document review tracking. In all these cases, data is extracted semantically in such a way that it can be further processed or analysed.
Much simpler handwriting recognition technology has existed since the early 90s (for instance, machines that read and processed cheques).
How does your technology move these capabilities forward?
SQ: At SearchInk we have realised throughout our research process that scaling the performance of handwriting recognition goes hand-in-hand with the context in which the text is written. If the context is understood, the capabilities and quality of the output increases exponentially.
Human beings sometimes can read words that are objectively distorted and unreadable — but because we recognize the context of the word, we are then able to “read” the relevant word or sentence.
For that reason, we have considerably invested in both content and context. In other words, we’re developing technology able to read like a human being. Interestingly this approach also resolves a far bigger problem that most companies have: extracting any type of data that is semantically connected, out of a high volume of highly arbitrary data sets.
The pairing of neural networks and deep learning is an incredibly exciting step.
In a general sense, how do you think this will affect our day-to-day lives?
SQ: Machine Learning and Deep Learning will significantly transform the way we work — but even more importantly, the way we live. In the medium and long term, almost all processes will be affected. At SearchInk, we decided, as a first step, to focus on specific use cases around clearly defined processes — however our aim and vision is to reach generalized AI within the next five years.
Further applications of this tech will improve our life on big scale by delivering ground-breaking solutions for industries like autonomous cars, health care, urban life or education.
If tech like Searchink’s can replace a lot of tedious, previously human-powered work, how will AI specifically affect our work experience?
Will we work less? Or will work be more fun?
SQ: We are at a key moment of work transformation. Past industrial revolutions made us produce more: it was about multiplying human forces to increase the outcome.
We’ve come to a place where we now need to make the outcome manageable for people. It’s about time to drive competences and skills towards part of the job that is really fulfilling.
Letting machine intelligence manage what is digested by other machines should give people space and freedom to be more creative.
People will always decide. Machine Intelligence is here to make room for faster and better decisions.
On a personal level, what specific application of this technology really excites you?
You’re clearly passionate about the product — what is it about it that gets you up every morning?
SQ: On a technical side, I am personally really excited by the fact that we are building something that is self-learning — which seems strange and unreal. But it reminds me of my years at school: the more you learn the better and cleverer you get — and now we adopt this approach into the technical world.
I’m also excited by the development pace. We’ll move soon from what we call “narrow AI” towards “generalized AI”, so transferring the learning from specific verticals and use cases into a generalized model that will have an impact not seen before.
On a social side — I am a strong believer that we are living in a very stimulating era, which one day in the future will be compared with the industrial or the internet revolution. What we are building here will have a substantial impact on the way we work and on what we are focusing on in our lives. This is really exciting. It is down to us to take this chance and define this game changing moment.
Don’t forget: SearchInk are hosting a huge satellite event on Friday 14th July — sign up for their Machine Intelligence Summit now!
Expect the unexpected: who should you look out for at TOA 2017? — we’ve just announced the 100th speaker — who are our team’s picks?
Will the cities of the future have an electric skateboard lane?––
Ryan Evans’ Inboard wants a future where you’ll be able to go wherever you want, whenever you want, safely, efficiently, and affordably
Originally published at blog.toa.berlin on June 8, 2017.
