AI is widespread and still growing rapidly, we write about it almost every day. But there are also disadvantages and risks, sometimes with potentially catastrophic consequences. To better investigate this, OpenAI is now doing more work itself.

Much (digital) ink has already been spilled about the potential dangers of AI to the world and human survival. From the godfather of technology to OpenAI CEO Sam Altman, who is also pondering massive job losses. The American company will now make greater efforts to assess the risks itself, including with a team set up specifically for this purpose.

A prepared team

It became the new OpenAI team Preparation team and is led by Aleksander Madry, professor at the famous MIT institute. The team will help track, assess, predict and protect against catastrophic risks.

Madry and his team work in four categories:

• Individual belief
• Internet security
• Chemical, Biological, Radiological and Nuclear Threats (CBRN)
• Autonomous Replication and Adaptation (ARA)

The mission of standby This also includes the development and maintenance of one Risk-informed development policy. This RDP should create a spectrum of protections and represent OpenAI’s approach to assessing and monitoring its AI models. Additionally, the policy must provide a structure of accountability and oversight of the entire development process.

Be prepared

To be even better prepared, OpenAI is also launching the Preparedness Challenge. The company relies on third parties to stay informed and continue to grow the team.

Anyone who wants can submit ideas to alert the team to less obvious areas where misuse of AI could have disastrous consequences.

Up to ten submissions will be rewarded up to $25,000 in API credits and the best ideas will be published. For those who have the ambition, the challenge is also about the team standby fill out.