Artificial intelligence can democratize access to health or education. This is what Pilar Manchón (Seville, 1972), director of AI research strategy at Google, believes, while defending that the time has come to use these tools to develop. … in the personal and professional sphere.
—The United States and China have successfully created generative AI tools. Why haven’t we achieved this in Europe?
“Europe has the potential to play a fundamental role. We are now behind the big leaders, particularly in the United States, but there is no reason not to invest more. Incentives must be created so that researchers and developers do not end up developing in third countries.
— Is AI regulated appropriately?
—At Google, we advocate for responsible regulation. This should be enough to mitigate risks and avoid problems, but without stifling innovation. Achieving perfect regulation the first time will be difficult for any country. Some will err on the side of excess, others on excess, but we will learn from each other.
— How to convince the skeptics?
—With data. If someone is closed-minded, arguments don’t matter because they won’t change their mind. But those who have the ability to reason can understand it. We have a great opportunity! It’s a tool that allows us to do more, faster, with better quality and gives us access to things we couldn’t even dream of before.
Anyone with virtually no training can do wonderful things. It’s a time of empowerment and dreaming, of thinking about how far you can go with tools that give you a brutal boost. We can build a wonderful world and guarantee access to education and health. The potential to become better people and better societies is brutal. Now, there are associated risks, of course. Any great tool with great potential can be used inappropriately.
Google Evolution
—How will Google’s AI tools evolve?
— Cover your ears, they’re going to throw me out! (she jokes with the Google communications colleague who accompanies her). It’s a very exciting time. We see where we are going and the potential that is beginning to awaken. There are people who understand that the quality of the “message” determines the quality of the system’s output. And also that not all models are the same: some are developed responsibly and are better protected than others!
The challenge now is to make those who already use artificial intelligence understand that it not only makes their lives easier, but also helps them develop, personally or professionally. Not only for daily tasks, but also to develop yourself, increase your abilities and become someone who can contribute more in your field or in society.
—Will the Google search engine as we know it die?
—Currently we prioritize the best answer and the user decides which one is most useful. The results abound in that many people prefer AI responses. But the beauty of this is that the system transforms and is bidirectional through usage and feedback. It’s an interesting process whose end we don’t yet know, but which seems very organic to me.
—In terms of total users, did ChatGPT beat its direct competitors, like Gemini, by being first to market?
—In the beginning there was a more experimental use of chatbots, but today much more specialized tools are used. Quality and safety are sought after. When the risks associated with certain tools are understood, Gemini is increasingly used.
—What sets Gemini apart from its competitors?
— Poof! How much time do we have? The fundamental difference is how it is done. It is not only the end goal that matters, but also the path to get there. We have been pioneers in the responsible development of our tools, taking into account the selection of data, the algorithms used, the tests carried out and the ethics committees that evaluate the updates. We never sacrifice quality or safety for speed! If I were on the other side of the fence, analyzing which system to use as the CEO of a company, I would not only look at quality, but also at security and innovation guarantees.
About The Author
