In the midst of the rise of Artificial intelligence, when in addition to happiness there are also worries, Google seeks to establish a retaining wall. that’s how it’s born SAIF, the enterprise security framework.
SAIF is an acronym for Secure AI Framework, and Google’s goal with it is to set “clear industry security standards to build and deploy technology responsibly.”
That’s the key word: responsibility.
According to Google, “SAIF is designed to help mitigate the specific risks of Artificial Intelligence systems, such as model theft, training data poisoning, malicious input injection, and sensitive information extraction.”
At present, governments are still in the phase of knowledge of AI, proposing legal frameworks. We have seen this with the European Union and the United States, which are looking for a Code of Conduct for developers, and now with China, according to Elon Musk after meeting with senior leaders of the country.
But companies want propose your own scheme, and Google represents a step forward.
Google SAIF’s main scheme against threats in Artificial Intelligence
The principles of SAIF include:
- Strengthen the security bases of the Artificial Intelligence ecosystem.
- Enlarge detection and response to bring Artificial Intelligence to an organization’s universe of threats.
- Automate defenses to keep up with new and existing threats.
- Harmonize platform level controls to ensure uniform security throughout the organization.
- adapt the controls to adjust mitigations and create faster feedback loops for AI implementation.
- and contextualize the risks of the AI system in surrounding business processes.
Phil Venables, head of information security at Google Cloud, told Axios: “Even as people look for the most advanced approaches, people really need to remember to get the basics right as well.”
The big question is what the measures will be like against direct rivals, as OpenAI, developer company of ChatGPT. This will undoubtedly be revealed over time.
But the first step is taken: Google launches at the forefront of security in the new era of Artificial Intelligence.