On 18th September, the Competition Market Authority published its long awaited report in AI Foundation Models.
There were some interesting chestnuts within the report, foremost amongst them the fact that the Authority seems to be following the British government’s line that AI has the potential to cause some serious damage to society.
Consequently, the report focuses heavily on the recommendation that to prevent customers and businesses being swept away by a few big companies, competition must be encouraged. By highlighting the possibility of AI tools being developed on the sly that could be used to perpetuate fraud or fake reviews, the Authority highlights the need for a healthy environment for AI to develop. With government encouragement if necessary.
On this last point, the report calls for considerations of safety, data protection and intellectual property rights. All things that Rishi Sunak, Britain’s Prime Minister, has said he would discuss at the upcoming World AI Conference that is to be held in the UK.
The report also goes onto highlight a few key points that it feels can help maintain competition and safety:
- Access: to things such as data, expertise and capital without restriction
- Diversity: both open and close sourced models
- Choice: a range of models from in-house, partnerships or plug-ins
- Flexibility: enabling consumers to switch and use multiple services easily
- Fair dealing: No anti-competitive conduct
- Transparency: Information available to allow for customers understanding of the tool
As AI continues to develop, it will be interesting to see whether this advice is followed and what it portends for the future.
Vivek Rajkhowa 
Leave a comment