BSA | The Software Alliance consistently engages with governments and institutions worldwide to shape the development of policies in a globally consistent manner. While much attention has been paid to policymaking in the European Union (EU) and the United States, here is a quick roundup of where – and how – BSA is engaging on tech policies worldwide.
AI Policy on the Move in Japan
The Government of Japan has advanced several key initiatives related to artificial intelligence (AI) in recent months, with more activity expected in the coming year.
- BSA continued to engage with Japan’s Ministry of Internal Affairs and Communications (MIC) and Ministry of Economy, Trade and Industry (METI) throughout a public consultation period on voluntary AI Guidelines for Business. BSA’s comments encouraged continued global harmonization around a risk-based approach to AI, a balanced allocation of responsibilities among actors in the AI ecosystem, and the protection of confidential and proprietary information.
- BSA also filed comments with the Agency of Cultural Affairs (ACA) seeking to interpret Japan’s Copyright Act as it relates to AI and copyright issues. BSA’s comments said that the current Copyright Act provides well-balanced guidance for the protection of copyrighted works. The Act also has the flexibility necessary to develop and produce new, useful services and works in the public interest. Read the full comments here.
Japan is likely to remain an area for focus in 2024, as some policymakers look to move forward AI legislation this year following the government’s establishment of a new AI Safety Institute.
Additional Governments Focus on AI and Copyright
BSA is also working in Korea and Canada to urge those governments to maintain fair use principles for training AI models.
- Korea: BSA told the Korea Copyright Commission that the non-consumptive computational analysis of content in AI training data sets constitutes “fair use” under Korea’s Copyright Act and encouraged multi-stakeholder efforts to minimize the risk of infringement.
- Canada: BSA filed comments to Innovation, Science and Economic Development Canada (ISED) stating that the computational analysis of copyrightable content in AI training data sets is not a copyright infringement under Canadian law, but also urged the development of a specific statutory exemption to the use of copyrighted material within AI training sets (to the extent that Canadian law is unsettled in this space).
BSA Urges Careful Approach to Deepfakes in India
BSA filed comments with India’s Ministry of Electronics and Information Technology (MeitY) as it considers amendments to the IT Act to address deepfakes. BSA urged MeitY to avoid a one-size-fits-all approach and recognize that many enterprise software products pose a limited risk and do not provide services directly to consumers. BSA additionally urged MeitY to look to content authentication tools like watermarks or standards advanced by the Content Authenticity Initiative (CAI) or the Coalition for Content Provenance and Authenticity to combat deepfakes.
On Cybersecurity, BSA Pushes for Clear Approaches and Commercial Software Use
Cybersecurity is another area of emphasis in global policy, as BSA engages with governments in the APAC region to pursue policies that provide clarity, and access to commercial software solutions.
- Singapore: As Singapore’s Cyber Security Agency (CSA) considers amendments to its 2018 Cybersecurity Act, BSA urged CSA to remove broad and duplicative sections imposing new rules for critical information infrastructure (CII) providers, and provide clearer criteria for identifying major Foundational Digital Infrastructure (FDI) and requirements for those entities. Find the full comments here.
- Australia: BSA urged the Australia Public Service (APS) to continue making use of commercial software tools and to avoid overly stringent sovereignty requirements for public procurement that would restrict growth of Australia’s tech sector, and instead align with internationally recognized standards.