IBM launches first AI incubator for social impact

Deadline extended! Apply for your organization now

By and Seth Dobrin | 6 minute read | July 12, 2021

Four people talking in office

We are proud to announce the launch of IBM’s first AI incubator for social impact organizations, empowering nonprofit organizations to explore and use high-impact data science and AI projects to advance their mission. The incubator, which will run from September 20 through December 10 2021, offers selected nonprofits opportunities to attend training workshops, hear from expert speakers, learn from each other’s experiences, and work with IBM AI leaders to implement a project during the 10-week program. Participating organizations will work alongside experts from IBM Data Science and AI Elite and IBM Research in an apprenticeship model, while leveraging the power of the IBM Cloud Pak for Data as a Service platform and building the support of a community of like-minded organizations. Ultimately, through the incubator and other AI for social good programs, we aim to support organizations with an identified gap or aspiration in data and AI, and help them enhance their long-term impact in the communities they serve.

The launch of the incubator comes in response to deep experience across IBM working with the social sector, and recognizes the immense opportunity demonstrated by inspiring examples of pioneering organizations already using advanced analytics and AI to accelerate their missions, and the significant gaps that exist within the sector at large in AI readiness, adoption, and understanding. IBM’s Institute for Business Value and IBM CSR conducted a global study of data and analytics in the nonprofit sector that highlighted gaps in talent, tooling, and ability to allocate internal budget.

Over the past several years, experiences through the long-running IBM Research Science for Social Good program demonstrate increased demand for advanced analytics in nonprofits, as organizations digitize at an accelerated rate due to declining costs, pandemic conditions, and sector demand. But while demand increases, barriers to adoption persist. Teams of IBMers from the Data Science and AI Elite, IBM Research, and Data and AI organizations worked alongside nonprofits on data and AI projects and reflected on the need for capacity building in the sector. They recognized the potential of a new model for working with organizations with apprenticeship and community at its core. Environments for responsible experimentation, mindful of fairness and unintended consequences with expert support and tooling for trustworthy AI, are scarce. These types of experiences are critical to enabling nonprofits to articulate the value of analytics.

“We’ve been observing an expansion of the digital divide across the social sector, and the data divide naturally follows,” said Becky Duane, Business Development Executive and AI for Social Good Program Lead. “Despite the challenges, many organizations are demonstrating the power of data and AI to enhance their missions, stakeholder engagement, and understanding of the communities they serve. We believe that capacity building through meaningful dedication of IBM tools and expertise, along with nurturing peer-to-peer collaboration, can empower these organizations that are forging the path for deeper data work the nonprofit sector.”

Collaborative projects with organizations like the Urban Institute and Change Machine facilitated the development of the incubator pilot. IBMers who lead engagements with enterprise customers spent time working with nonprofit organizations on social-sector-focused problems, then came together to craft the incubator program designed to address this gap. Andre Violante, an IBM Data Science and AI Elite manager, highlighted the grassroots nature of the program: “It came together organically with passionate and creative people across Data and AI and IBM Research. Brainstorming was mutually educational with the depth of experience across social and nonprofit sector, tech startup experience, and the traditional engagements across a variety of industries. We recognized the power and potential of a more scalable and collaborative partnership with social impact organizations, which would empower them to take their data science and AI work to the next level.”

AI leaders from IBM’s Data Science and AI Elite teamed up with the Urban Institute’s Chief Data Scientist to build and improve upon models to predict neighborhood change, helping to inform policymakers on areas of opportunity for support and intervention before the change would otherwise be recognized. “Working with IBM’s Data Science and AI Elite team enabled us to take advantage of the latest advances in data science, expose Urban’s team to new techniques, tools and ways of working, and contribute a real advancement to the field all at once,” said Graham MacDonald, Chief Data Scientist at the Urban Institute. “In my experience, these types of engagements are valuable not only to making impact with the tools we build, but more importantly, allow nonprofits like Urban to build capacity to accelerate change and close the data skills divide.”

The team also partnered with Change Machine, a national nonprofit building financial security through people-powered technology, in an apprenticeship model to develop a recommendation engine to ensure that appropriate, safe and non-predatory financial products are offered to customers in low-income neighborhoods.

“The engagement with IBM’s AI for Social Good team taught us how to leverage our data in new ways and how to build a framework for creating and managing machine learning models,” said David Bautista, Director of Product at Change Machine. “The project also served as a springboard for our involvement in advanced cloud solutions and helped us deploy a real-world application of AI, something we previously considered to be years away at best.”

Any US-based nonprofit organization can apply here by providing information on current practices and a proposed project for the incubator. Organizations may submit project ideas whose relevant data is either proprietary or public, structured or unstructured. It is important that the organization have access and permission to use the data in a project by the incubator start date. Projects should ideally be focused on an existing data gap that augmented expertise, guidance, and tooling can begin to address in an 8-week period for active project work during the 10-week overall program. Applications will be reviewed on a rolling basis and accepted through August 13. Finalists will be announced by August 20, and the cohort will be announced August 31.

Program details

Timing

  • Application Deadline (EXTENDED): August 13, 8pm EDT
  • Selection: Rolling review
  • Finalist Announcement: August 20
  • Cohort Announcement: August 31
  • Program: September 20 – December 10

Benefits through:

Community

  • Cohort of social sector organizations
  • Future access to alumni community and advisors
  • Blog and media opportunities to showcase work with IBM

IBM will provide:

Expertise
Trainings and sessions featuring IBM experts from Research, Data and AI, Design, and Ethics, and a deep dive project with dedicated data scientists from IBM’s Data Science and AI Elite team and IBM Research advisors.

Tools and cloud platform

Team platform on Cloud Pak for Data as a Service during the incubator, and access to Cloud Pak for Data as a Service with credits post-incubator

Program structure

  • Launch week: speakers, working sessions, onboarding and trainings
  • Project sprints with Data Science and AI Elite: four two-week sprints with your team working alongside IBM data scientists and sharing lessons learned with cohort
  • Final presentations and readouts: speakers and project readouts for the full cohort

Who we’re looking for

We are seeking applications from nonprofit organizations that have identified a use case that could leverage advanced analytics or AI to amplify their social mission. While we will seek to create a cohort that has a diversity of sophistication of data and AI use in their organization, applicants that are best suited for the program are those that have:

  • Some ideas on potential use cases
  • One or two (or more!) individuals who can engage and work with data scientists based on basic programming or statistics experience, and who can carry the project forward post incubator
  • Access to data or ideas of publicly available data that can be used during the incubator.
  • Organizations willing to share high level problem statements, approach, and learnings with other nonprofit organizations in the cohort (with flexibility depending on confidential nature of the use case) 

For more information, contact ai4good@us.ibm.com.

IBM is committed to advancing tech-for-good efforts, dedicating IBM tools and skills to work on the toughest societal challenges. Building on existing programs such as the IBM Science for Social Good initiative, the IBM Data and AI team is helping nonprofit organizations accelerate their mission and impact by applying data science and machine learning approaches to social impact use cases.