AI in healthcare:

Concerns about AI’s existential risks have garnered attention from prominent figures in science and technology. Physicist Stephen Hawking warned in 2014 that AI could signal humanity's demise. Elon Musk has likened its dangers to those of nuclear weapons, advocating for Mars colonisation as a safeguard against a potential rogue AI. Microsoft co-founder Bill Gates has echoed these warnings, aligning with Musk’s apprehensions about superintelligence. Even Sam Altman, CEO of OpenAI, the organisation behind ChatGPT, has publicly acknowledged the possibility of AI-driven global upheaval.

In 2023, anxiety over AI's unchecked growth culminated in a bold proposal by the Future of Life Institute and the Centre for AI Safety. These organisations called for a six-month pause on developing systems more advanced than OpenAI's GPT-4, equating AI risks to those of nuclear war and pandemics. Their appeal, backed by industry leaders, reflected fears about humanity’s ability to manage technology's rapid development responsibly.

Who shapes AI?

The debate over AI’s future extends to its ethical oversight. Philosopher Nick Bostrom’s concept of "existential risk" (X-risk) highlights threats that could annihilate humanity or permanently hinder its potential. However, ethical discussions around AI are increasingly dominated by corporate interests. The 2023 Stanford AI Index Report revealed that 71% of AI ethics publications now come from industry, a significant shift from academia's historically dominant role.

This corporate influence poses a critical question: Can private companies, driven by profit motives, fairly shape AI’s ethical frameworks? These entities often prioritise the perspectives of high-income nations, risking the exclusion of voices from the Global South. This disparity raises concerns about AI reinforcing global inequalities, particularly when algorithms designed in one context are applied to vastly different socio-economic settings.

Everyday risks of AI

While existential risks dominate headlines, more immediate concerns about AI demand attention. Algorithmic bias, AI hallucinations, misinformation, and privacy breaches are pressing issues. In medicine, for instance, studies suggest GPT-based systems may offer biased diagnoses based on race or gender. Additionally, the digital divide exacerbates inequality, as data from high-income regions enrich AI algorithms, leaving low-income populations underrepresented and underserved.

Other ethical concerns include the displacement of creative work and the potential misuse of AI in surveillance or misinformation campaigns. For AI to be implemented responsibly, it is crucial to address these risks alongside the broader existential debates.

Despite its risks, AI has already demonstrated its ability to transform fields like healthcare. In drug discovery, DeepMind’s AlphaFold revolutionised protein structure prediction, accelerating research into diseases and treatments. AI-assisted diagnostics have shown promise in identifying anomalies in radiology images and improving disease detection accuracy.

In neuroscience, AI has been instrumental in decoding brain activity, offering new hope to individuals with communication or mobility challenges. Additionally, AI-driven tools have contributed to combating antimicrobial resistance, leading to the discovery of new antibiotics.

Public perception

Public attitudes toward AI vary significantly by region. Anglophone countries like the US and UK often express anxiety over AI, whereas nations such as Japan and South Korea exhibit greater optimism. These differences may stem from contrasting regulatory frameworks, cultural attitudes, and levels of trust in technology.

To address AI’s risks proportionately, it is essential to recognize that other existential threats—climate change, nuclear weapons, and antimicrobial resistance—are equally pressing. A united global effort to tackle shared challenges could reduce AI-related risks and foster intergenerational trust.