MIT Sloan Research on Artificial Intelligence and Machine Learning
There is no doubt that artificial intelligence and machine learning are playing an increased role in business decision making. A 2022 survey of senior data and technology executives by NewVantage Partners found that 92% of large companies said they saw returns on their data and AI investments, up from 48% in 2017.
But as these technologies enter the mainstream, new questions arise: how will they change the nature of workflow and connection to the workplace? Will they be ethically operated? Will they replace humans?
Here’s what to consider as AI and machine learning become ubiquitous, according to MIT Sloan researchers, visiting scholars and industry experts.
Artificial intelligence is disrupting most professions, but it is far from replacing humansaccording to a book examining the findings of the MIT Task Force on the Work of the Future.
Some 92% of large enterprises report a return on investment in data and artificial intelligence.
MIT researchers David Autor, David Mindell, and Elisabeth B. Reynolds argue that understanding the capabilities and limitations of artificial intelligence is critical when thinking about its impact on jobs.
Current AI challenges center on physical dexterity, social interaction, and judgment. Consider the example of a home health aide, whose responsibilities include providing physical assistance to a fragile human being, observing their behavior, and communicating with family and doctors. Only when automation reaches this level can it truly be considered what researchers call “artificial general intelligence.”
It is possible to harness AI to create a more equitable future. Across industries, workers fear that automation and artificial intelligence will steal their jobs. Professor of Management at MIT Sloanshares these concerns, but also sees “enormous” potential for innovation in new technologies to create “a productive and more equitable future”.
In his online executive education course, “Leading the Future of Work,” Kochan lays out a four-pronged roadmap for the workplace of the future:
- Strive to become a “high road” enterprise that creates value for all stakeholders, including their employees.
- Use advanced technology to drive innovation and increase work.
- Train and develop the workforce to work with new technologies.
- Rebuilding a dialogue with union leaders for a mutually beneficial future.
An AI-powered productivity boom is on the way. Consider the Internet: its foundational technologies took root in the 1960s and 1970s, but weren’t commercialized until the mid-1990s. the work of the future, calls this phenomenon a “J-curve”, when technological acceptance is “slow and gradual at first, then accelerates to reach broad acceptance”.
Now, companies must prepare for an AI-powered J-Curve as the technology takes off. Businesses should focus on integrating artificial intelligence and machine learning into work processes and employee readiness, Brynjolfsson told an EmTech Next conference, while policymakers should act to ensure that its adoption does not contribute to inequalities.
AI requires stakeholder buy-in. Machine learning tools are used in many fields. But bringing technology into the workplace is only one step – these tools only succeed if they’re integrated into workflows and people trust them enough to depend on them.
According to a study by MIT Professor Sloan, the key to successful adoption is an ongoing dialogue between technology developers and end users.and co-authors.
“Managers and developers must engage in a back-and-forth process to create, evaluate, and refine tools so they’re useful in practice,” Kellogg said.
Additionally, stakeholders must believe that AI programs are accurate and trustworthy. The explainability of artificial intelligence can help. Researchers at the MIT Center for Information Systems Research define AI explainability as “the ability to manage AI initiatives in a way that ensures models generate value, are consistent, representative, and trustworthy.”
The explainability of artificial intelligence is an emerging field, admit the researchers. They recommend that companies start by identifying units and organizations that are already creating effective explanations of AI, and identifying practices that the organization’s own AI project teams have already adopted.
AI and machine learning are transforming digital marketing. Most marketers are concerned about retention and revenue, but without good forecasts, decisions about effective marketing interventions can be arbitrary, saidHead of the Social and Digital Experimentation Research Group at MIT’s Digital Economy Initiative.
Machine learning will change that, helping to predict customer behavior and understand their needs.
“There’s a lot of value in applying statistical machine learning to predict long-term, hard-to-measure outcomes,” Eckles said.
Companies like Wayfair and Spotify are leveraging machine learning for tailored customer experiences, from highly personalized furniture search results to personalized suggested playlists. And, as COVID-19 spread, Moderna used its longstanding automated processes and AI algorithms to scale up the number of small-scale messenger RNAs (mRNAs) needed to conduct clinical experiments. This groundwork contributed to Moderna’s publication of one of the first COVID-19 vaccines (using mRNA) early in the pandemic.
Good data makes good AI. Having five essential data capabilities, such as data scientists and a data platform, helps companies build successful AI programs.
According to researchers at MIT CISR, the key is to adopt a corporate rather than a local perspective as AI project teams learn and mature. When companies identify and accumulate the expertise and practices of their AI teams, they can create reusable and tweakable practices and build their capabilities. This accelerates new AI projects and prepares these future teams for success.
Embracing data-centric AI is also important. According to Andrew Ng, SM ’98, founder of the Google Brain research lab and former chief scientist of Baidu, it is “the discipline of systematic data engineering needed to build a successful AI system. “.
Focusing on high-quality data that is consistently labeled could unlock the value of AI for industries such as healthcare, government technology and manufacturing, Ng told an EmTech Digital conference.
And while AI can’t solve all the problems in the world — and might even cause new ones along the way — it can at least make Wordle a little easier to solve.
Read next: SEC’s Gary Gensler explains how AI is changing finance