The Innovation Workshop on the second day of the Global IndiaAI Summit 2024 sparked a stimulating discussion on AI safety, democratizing AI access, and ensuring sustainability. 

With national and international participants, the workshop highlighted the crucial need for the global community to tackle the primary challenges of AI safety, making advanced AI technologies accessible to everyone while promoting sustainable practices.

The entire international community needs to address the main challenges of AI safety, as demonstrated by the Innovation Workshop, which saw national and global participation.

Continuing the progress of the Innovation Workshop on AI, this year's innovation workshops focused on AI safety, democratisation of AI Resources and Environment, and sustainable individuals.

“The idea was to turn the discussion into action by creating this unique platform with all members. As you facilitate singular discussions and feedback, in this process, we were able to refine the shape of the two tracks we had defined at the beginning, and it will often identify priorities.

One of the deliberation points was the need for adequate protocols and legal frameworks for SMEs. Risk assessment and IP protection regarding democratisation were also on the agenda.

“We identified the priority of our local stakeholders, making sure that AI is multilingual and multicultural, which often goes together, and that the data is accessible in local languages,” per the panel.

These were through open-source and environmentally sustainable agriculture via solidarity over international collaboration, which also considers local specificity and investment in research and development. They added that tenability must also be profitable, as we're seeing, and to find adequate financing programs.

Labelling safety as a broad term, the GPAI projects from Serbia and the UK were about fair work.

The goal was to improve the working conditions of data and temp workers in the AI value and supply chain by scaling up assessment methodologies and the certification process and applying AI literacy and knowledge awareness.

The final project of the safety track, which had ten selections, dealt with ideas to create conditions for governments to spread knowledge and use AI safety to all citizens. The goal was to provide governments with the right tools, processes, methodologies, and resources to spread this knowledge and use of AI safety to all citizens.

The democratisation of AI, or AI for all, presents the reality of AI opportunity being equitable, wherein everyone can access it either as a user or as a developer, illuminator, creator or educator.

The deliberation, which also included discussions from the Singapore delegation, moved on to data sharing for public good, such as in use cases for curing diseases like COVID-19.

“I think we're all realizing that there is no option. We do need to think of some form of regulation. We were supposed to come up with something that will form a basis for all of us to consider, and similarly for data that one needs to provide, versus what data will be shared and how it will be shared, and what format it will be shared in,” the panel deliberated.

As part of the 2025 work plan, the interested nations in the working group, partners like the National Air Safety Institute and National Institute, and other international bodies, organisations, and initiatives deliberated on what is needed more for education and upskilling and the need to reach a broader community and audience.

“In terms of stakeholders, we want to remain very agile. Regarding what types of partnerships we can build around the projects, and depending on the different stakeholders, it can be like civil society. Private organizations like in the project recorded with pets, the privacy-enhancing technology, we partnered with a company in Singapore and the government, so it means every stakeholder should be included and could be included in those partnerships around delivering the most impactful projects,” they noted.

The integration workshop approach is hands-on and not a conference. Experts and audiences came together to deliberate on the future of AI and the need for regulations to save jobs and the creator economy. They concurred that AI was more than a market or business issue; it was social.

“After the Montreal conference, we didn't realize how much work it would mean, so it is an intense methodology that can provide roadmaps quickly for applied and responsible AI solutions,” the panel concluded.