IBM Leading in AI Innovations: What the Company is Working on Now

IBM has been a leader in computer and IT innovations since 1945. Artificial Intelligence is big news today, and of course, IBM is still a prominent player in the field. After all, who can forget Deep Blue, IBM’s AI from the 1990s, and its chess match against world chess champion Garry Kasparov?

Naturally, the tech world is curious about what IBM is up to today, specifically in the field of Artificial Intelligence. So, what is the foremost computer leader doing with one of the hottest technologies, AI innovations, of the century? Join us now as we pull back the curtain on IBM projects currently in the works.

Artificial Intelligence Engineer

Your Gateway to Becoming a Successful AI ExpertView Course
Artificial Intelligence Engineer

The Latest AI IBM Projects

Here is a list of IBM’s current projects relating to AI innovation. Some of them started last year and are currently ongoing. However, these research projects aren’t the only moves IBM is making to strengthen its role in AI innovation. IBM plans on purchasing Waeg, a Salesforce consulting partner. This move extends its range of hybrid cloud services and supports its Artificial Intelligence strategy.

This entry is more of a series of indirect projects involving the MIT-IBM Watson AI Lab. However, it’s a significant venture, considering the clear and present danger that COVID still poses. These projects range from early sepsis detection in COVID-19 patients to using repurposed drugs to treat COVID to bringing the world back to normal with mass testing, targeted lockdowns, and personalized treatments.

This AI innovation project is tasked with improving research on AI agents that can control a simulated mobile robot guided by computer vision to manipulate objects, pick them up, and carry them to new locations. The agent must locate small objects distributed around the simulated house, pick them up, and take them to a new, final spot.

In the quest for human-style comprehension in symbolic AI systems, scientists are working on new hybrid systems. This AI innovation involves combining neural networks that pull statistical structures from raw data (e.g., image files), with symbolic representations of logic and problems. Ultimately, researchers hope to make machines learn more like people do—mastering abstract concepts and connecting words with images. Researchers are working on giving machines actual common sense and reasoning capabilities.

Only by trusting in AI innovations can the technology spread through the public. People fear what they don’t know and consequently are reluctant to embrace change and rely on new technology. IBM is developing a series of toolboxes and toolkits designed to inspire confidence and trust by explaining how machine learning models make predictions. Furthermore, the toolboxes address issues such as bias and discrimination. The keywords here are robustness, explainability, and fairness.

Artificial Intelligence requires breathtaking speeds and tremendous processing power to function at its full potential. Consequently, IBM’s hardware research teams are tackling this challenging task. Researchers are working on new architectures, devices, and algorithms to improve processing efficiency and facilitate the transition between Narrow AI to Broad AI. Researchers are using approaches such as in-memory computing, approximate computing, quantum computing, and machine intelligence. Specific examples include Digital AI Cores, Analog AI Cores, Heterogeneous Integration, and AI Optimized Systems.

IBM research teams are developing tools designed to help Artificial Intelligence engineers spend less time training, updating, and maintaining their models. One big hurdle in the path of widespread AI acceptance is the difficulty organizations experience in trying to implement AI in real-world settings. Researchers are working on a concept called Automated Machine Learning (or AutoML), which automatically configures machine learning pipelines. This AI innovation means that non-ML professionals could build and use machine learning systems without needing extensive knowledge of the subject. By demystifying the concept of AI and easing its implementation, acceptance and use will increase worldwide. Put in plainer language, researchers are looking into ways of automating automation!

If you take these research projects as an aggregate, you can see where IBM is going this year. They want to see more significant AI innovation and implementation worldwide, and they’re using a multi-pronged approach. First, they’re developing improved ways for computers to think, reason, and react like people. Second, they are making the technology more straightforward for everyone to understand and use, regardless of their expertise (or lack thereof) in the field of AI and Machine Learning. Third and finally, they are working towards building greater trust and confidence in the new tech.

FREE Course: Intro to AI

Learn the Core AI Concepts and Key SkillsStart Learning
FREE Course: Intro to AI

The Watson Factor

Watson is IBM’s question-and-answer computer system, made up of a portfolio of business-ready applications, solutions, and tools. Watson was created to reduce costs and AI obstacles while optimizing the responsible use of AI and its outcomes. Watson helps organizations automate complex processes, predict future outcomes, and optimize employees’ time.

IBM researchers have been developing new innovative capabilities for Watson, focusing on automation, natural language processing, and trust-building, while infusing more intelligence into workflows. IBM has recently announced planned research into new Watson capabilities, such as the IBM Watson Discovery, which identifies more precise answers to natural language questions, including a score showing system confidence in the provided answer. Additionally, IBM Watson Assistant’s search skill is in beta, which automates Q&A pair extraction from FAQ documents, keeping a company’s virtual assistants up to date and decreasing the time-consuming work of manually updating the Q&A system.

Master Deep Learning, Machine Learning, and other programming languages with Artificial Intelligence Engineer Master’s Program

Would You Like a Career in AI Innovation?

The field of Artificial Intelligence is ripe with fantastic potential and growth, and you can get in on this rapidly developing industry. Simplilearn can help you get your start with its Artificial Intelligence Engineer Master’s program. Run in collaboration with IBM, the course features exclusive IBM hackathons, masterclasses, and “Ask me anything” sessions. You will learn data science concepts with Python, machine learning, deep learning, and NLP — through live, online sessions, self-paced instruction, practical labs, and real-world projects.

Indeed has AI engineers in the United States earning an average base salary of USD 149,184 per year. Glassdoor reports that AI engineers in India can make an annual average of ₹ 850,396.

Check out Simplilearn today and get your start on the ground floor of a field representing the future of IT in the 21st century!

About the Author


Simplilearn is one of the world’s leading providers of online training for Digital Marketing, Cloud Computing, Project Management, Data Science, IT, Software Development, and many other emerging technologies.

View More
  • Disclaimer
  • PMP, PMI, PMBOK, CAPM, PgMP, PfMP, ACP, PBA, RMP, SP, and OPM3 are registered marks of the Project Management Institute, Inc.