{"id":25,"date":"2026-07-21T06:30:47","date_gmt":"2026-07-21T05:30:47","guid":{"rendered":"https:\/\/blogs.bath.ac.uk\/idsb\/?p=25"},"modified":"2026-07-21T07:49:17","modified_gmt":"2026-07-21T06:49:17","slug":"why-safe-and-responsible-ai-matters-more-than-ever","status":"publish","type":"post","link":"https:\/\/blogs.bath.ac.uk\/idsb\/2026\/07\/21\/why-safe-and-responsible-ai-matters-more-than-ever\/","title":{"rendered":"Why safe and responsible AI matters more than ever"},"content":{"rendered":"

By Emma Walker, Andre Barrinha and Julian Padget.<\/strong>\u00a0<\/span><\/p>\n

Artificial intelligence is developing rapidly, bringing significant opportunities alongside new risks. As governments, researchers and industry consider how best to govern these technologies, ensuring AI is developed and used safely, ethically and inclusively has become an increasingly urgent challenge.<\/span><\/i>\u00a0<\/span><\/p>\n

These issues were at the centre of the United Nations Global Dialogue on AI Governance<\/a>,\u00a0held\u00a0in Geneva\u00a0in July, where IDSB Deputy Director Dr Julian Padget\u00a0represented\u00a0the Institute following a written submission highlighting why effective AI governance matters. Drawing on those discussions, and the\u00a0expertise\u00a0of researchers from across the Institute, this blog explores why safe and responsible AI matters more than ever.<\/span><\/i>\u00a0<\/span><\/p>\n

Real-world\u00a0risks\u00a0are\u00a0already\u00a0emerging<\/span><\/b>\u00a0<\/span><\/p>\n

In recent years,\u00a0we have seen\u00a0examples\u00a0where potential harms related to AI systems have not\u00a0been\u00a0identified\u00a0or\u00a0addressed before\u00a0technologies have been\u00a0deployed. For example,\u00a0<\/span>Grok AI\u00a0was used\u00a0to create demeaning images of others<\/span><\/a>, primarily targeting\u00a0women, with action only\u00a0being\u00a0taken\u00a0following widespread backlash.\u00a0Generative\u00a0AI has also been\u00a0used\u00a0to\u00a0influence public opinion on a range of issues through the\u00a0<\/span>creation and circulation\u00a0of fake information<\/span><\/a>\u00a0<\/span>or to fraudulently extract money or information from individuals and\u00a0organisations via\u00a0targeted\u00a0<\/span>family emergency scams<\/span><\/a>\u00a0and\u00a0<\/span>CEO deepfake scams<\/span><\/a>. Large language models\u00a0were\u00a0found to\u00a0<\/span>expose users to extremist content and politicised communities\u00a0during conversational interactions<\/span><\/a>,\u00a0essentially\u00a0bypassing\u00a0safeguards\u00a0designed\u00a0to reduce\u00a0potential\u00a0risks\u00a0of harm.\u00a0<\/span>\u00a0<\/span><\/p>\n

And\u00a0it is\u00a0not just the way that we communicate and\u00a0interact\u00a0online\u00a0that is being\u00a0impacted. Generative AI\u00a0also presents a cybersecurity risk,\u00a0<\/span>aiding\u00a0the development of ransomware and\u00a0even\u00a0enabling\u00a0sophisticated cyberattack<\/span>s<\/span><\/a>. Although some of these\u00a0techniques existed\u00a0before\u00a0the\u00a0widespread rollout of AI,\u00a0<\/span>the added realism, ease of access and ease of creation that generative AI provides<\/span><\/a>\u00a0means that the scale and impact of\u00a0these\u00a0threats\u00a0have\u00a0grown substantially.\u00a0<\/span>\u00a0<\/span><\/p>\n

The\u00a0need for\u00a0collaboration and\u00a0governance<\/span><\/b>\u00a0<\/span><\/p>\n

There are examples of industry\u00a0collaboration\u00a0to reduce AI-related risks in specific areas, as seen in the Anthropic-led initiative\u00a0<\/span>Project Glasswing<\/span><\/a>. However, these approaches must involve a wider range of voices, including government, civil society and academia. This will enable a <\/span>fuller\u00a0consideration of potential\u00a0impacts at\u00a0the individual, group and societal level<\/span><\/a>s<\/span>.<\/span>\u00a0<\/span>\u00a0<\/span><\/p>\n

Most importantly,\u00a0governance\u00a0approaches need to be interoperable across regions, cultures, providers, deployers and harm types. They should\u00a0also\u00a0provide\u00a0sufficient clarity to minimise\u00a0<\/span>so-called\u00a0\u2018avoision\u2019\u00a0behaviours<\/span><\/a> where organisations may attempt to operate in the grey area between avoidance and evasion. This is vital if potential gaps between these various areas, and the risks they create, are to be avoided. <\/span>\u00a0<\/span><\/p>\n

Although\u00a0<\/span>there can be political reluctance to regulate AI to avoid stifling innovation<\/span><\/a>,\u00a0we need to\u00a0consider how these challenges can be overcome most effectively. For instance,\u00a0<\/span>is it feasible for individual regulators to oversee AI use within their respective\u00a0sectors<\/span><\/a>\u00a0rather than implementing overarching,\u00a0sector-agnostic regulation?\u00a0<\/span>\u00a0<\/span><\/p>\n

It\u2019s\u00a0not\u00a0enough\u00a0for\u00a0technologies with\u00a0the\u00a0potential to\u00a0transform\u00a0society,\u00a0law\u00a0and order\u00a0to be\u00a0addressed only\u00a0through isolated\u00a0use cases, disparate initiatives,\u00a0or self-regulatory\u00a0approaches that are not fully transparent,\u00a0open\u00a0or collaborative in approach.\u00a0<\/span>\u00a0<\/span><\/p>\n

Building a\u00a0safer AI\u00a0future<\/span><\/b>\u00a0<\/span><\/p>\n

Fundamentally, AI\u2019s potential to\u00a0support\u00a0complex\u00a0decision-making,\u00a0assist\u00a0with\u00a0task completion, and augment human capability can only be realised if it is developed in line with effective\u00a0<\/span>ethical and responsible innovation practices<\/span><\/a>. This is\u00a0currently hindered by uncertainties regarding\u00a0appropriate governance\u00a0mechanisms across multiple areas of application and\u00a0the need for governance frameworks that are\u00a0sufficiently flexible and rapidly adaptable, to accommodate new capabilities and novel\u00a0forms of\u00a0(mis)use.\u00a0<\/span>\u00a0<\/span><\/p>\n

Going forward, it is crucial that appropriate mechanisms for ethical and responsible governance of both AI development and AI use, including how these mechanisms may change and evolve, are developed and adapted collaboratively over time with a full range of stakeholders. This includes industry partners, academic researchers, policy specialists, end-user communities and those representing the natural world and other non-human systems that may be affected by AI.<\/span>\u00a0<\/span><\/p>\n

\u00a0One thing is\u00a0clear,\u00a0<\/span>the future impact of AI will depend on the choices we make today. By prioritising responsible innovation,\u00a0appropriate governance\u00a0and ongoing collaboration, we can ensure that AI serves as a force for good\u00a0-\u00a0enhancing our capabilities while protecting the wellbeing of individuals and\u00a0society as a whole.<\/span>\u00a0<\/span><\/p>\n

At the Bath Institute for Digital Security and Behaviour, we continue to develop our work in this key area.\u00a0With\u00a0expertise\u00a0spanning\u00a0policy, computer science,\u00a0psychology\u00a0and management, the IDSB takes a multidisciplinary approach to understanding AI that has much to contribute to global discussions on this topic.\u00a0<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

By Emma Walker, Andre Barrinha and Julian Padget.\u00a0 Artificial intelligence is developing rapidly, bringing significant opportunities alongside new risks. As governments, researchers and industry consider how best to govern these technologies, ensuring AI is developed and used safely, ethically and...<\/p>\n","protected":false},"author":2077,"featured_media":26,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":"","jetpack_post_was_ever_published":true},"categories":[9,10,8,4],"tags":[],"class_list":["post-25","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-artificial-intelligence","category-cybersecurity","category-digital-technologies","category-extremism"],"acf":[],"jetpack_featured_media_url":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-content\/uploads\/sites\/239\/2026\/07\/banner_dialogue.jpg","_links":{"self":[{"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/posts\/25","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/users\/2077"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/comments?post=25"}],"version-history":[{"count":0,"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/posts\/25\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/media\/26"}],"wp:attachment":[{"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/media?parent=25"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/categories?post=25"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.bath.ac.uk\/idsb\/wp-json\/wp\/v2\/tags?post=25"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}