IBM's 5 Technology Predictions for the Next 5 Years
IBM's 5 Technology Predictions for the Next 5 Years
At the IBM Think conference, IBM will release its annual five technology predictions for the next five years. Here's an insider look.
Join the DZone community and get the full member experience.Join For Free
Bias comes in a variety of forms, all of them potentially damaging to the efficacy of your ML algorithm. Read how Alegion's Chief Data Scientist discusses the source of most headlines about AI failures here.
IBM has merged all of its top conferences into one this year for IBM Think, a global event that 40,000 tech aficionados are expected to attend from March 19-22, 2018 in Las Vegas. The conference will cover topics like artificial intelligence, data analytics, and IoT and aims to "make the world of business work smarter."
IBM Think will answer questions about the new technological innovations that are coming soon and how these innovations will impact our lives. At the event, IBM will release its annual five technology predictions for the next five years. These predictions involve crypto-anchors, quantum computing, hacking, AI bias, and — wait for it — small, autonomous AI microscopes.
Nobody likes knockoffs. Crypto-anchors and blockchain will unite against counterfeiters. Fraud costs the global economy more than $600 billion every year, but crypto-anchors can make things complicated for bad actors. These tamper-proof digital fingerprints developed by IBM can be embedded in products and linked to the blockchain to help prove authenticity.
For example, crypto-anchors can help prove that life-saving drugs are legitimate and not counterfeit. Individual malaria pills can be coated in an edible shade of magnetic ink, and with a simple scan of a smartphone, a doctor or patient can immediately identify that the pill is safe and genuine. There is also a crypto-anchor that combines special optical devices with AI algorithms to identify the structure and labels of objects to verify that they're really what they say they are — for example, a 1982 bottle of Bordeaux or an expensive metal.
IBM has also designed the world's smallest computer: an edge device architecture and computing platform that is smaller than a grain of salt, costs less than ten cents to manufacture, and can monitor, analyze, communicate, and act on data.
The first models of these crypto-anchors could be available as soon as the next 18 months. Within the next five years, advances in various fields will take these systems into the marketplace.
2. Quantum Computing
Today, quantum computing is a researcher's playground. In five years, it will be mainstream, helping new categories of professionals and developers solve problems once considered unsolvable. According to IBM, knowledge of quantum computing will become a prerequisite for science and engineering programs, and universities will have quantum computing programs that teach students how to run real, practical experiments on quantum computing through the cloud. Quantum algorithms will be taught alongside classical algorithms.
These advancements will initiate the dawn of the commercial quantum era, where quantum computers will be able to simulate larger molecules, chemical reactions, and atomic bonding. This will lead to the creation of novel materials, development of more personalized drugs, and discovery of more efficient and sustainable energy sources.
In the future, quantum computers will no longer be seen as mysterious and will be embraced by the general public. Within the next five years, the industry will have discovered the first applications where a quantum computer can be used alongside a classical computer to solve specific problems.
Hackers gonna hack — until they encounter lattice cryptography. Many years from now, a quantum computer will be able to quickly sift through all the possible security probabilities and decrypt even the strongest encryptions. IBM is developing a new security method built on lattice cryptography, which hides data inside complex algebraic structures called lattices.
In the field of mathematics, lattices present problems that are very hard to solve. This difficulty is useful for cryptographers because it can be applied to protect information, even when hackers have quantum computing on their side. Lattice-based cryptography is also the basis for fully homomorphic encryption (FHE), which makes it possible to perform calculations on a file without ever seeing sensitive data, preventing it from being exposed to hackers. With FHE, credit reporting agencies can analyze and produce credit scores without decrypting personal data, and primary care physicians can share medical records with pertinent data without revealing a patient's identity.
Five years from now, our current security measures will seem woefully inadequate.
4. AI Bias
AI bias will explode in the next five years. AI systems are only as good as the data we put in them, and if we input data with certain racial, gender, or ideological biases, then the AI will be trained to be prejudiced. Unfortunately, many AI systems are currently trained with this bad data. IBM researchers are thinking about how to ensure that human bias does not affect AI.
The MIT-IBM Watson AI Lab is looking at using computational cognitive modeling to build machines that apply certain human values and principles in decision-making — and a crucial principle is to avoid bias and discrimination. IBM researchers have developed AI algorithms that reduce bias in training data by learning from previous data that has been recognized to be discriminatory. IBM scientists are also developing an unbiased AI service that can a) compensate for data bias, b) follow the bias in the training set, or c) introduce bias. The end user can determine the trustworthiness and level of bias in each scenario.
Within the next five years, reducing bias in AI systems will be key to humans trusting AI — and only unbiased AI will survive.
5. AI Microscopes
Our oceans are dirty. AI-powered robot microscopes may save them. In our current state, by 2025, more than half of the world's population will be living in water-stressed areas — and scientists struggle to collect the relevant data to prevent this. Even specialized sensors that can detect water conditions and chemicals aren't trained to anticipate unexpected scenarios.
In particular, detecting plankton is an effective way to sense aquatic health. However, studying these creatures is difficult because of their microscopic size. IBM researchers are building small, autonomous, AI-powered robot microscopes that can monitor plankton, better understand plankton behavior and how they respond to environmental changes, and predict threats to our water supply. The microscopes have chips that capture the shadow of the plankton to generate a digital sample of its health, without the need for focusing.
Within the next five years, these microscopes will be advanced enough to analyze interpret data locally and in real-time.
Opinions expressed by DZone contributors are their own.