"Ladies and gentlemen,
It is a pleasure to be with you today. Many thanks for inviting me to San Francisco.
When it comes to cyber-attacks, you could say my experience is somehow special.
Estonia is a small Baltic nation on the edge of Europe. We not only share a border with Russia, we also have a long and difficult history together.
In 2007, I was Prime Minister of Estonia. Over three weeks, my country was the target of an orchestrated cyber-campaign to destabilise parts of our online presence and civilian infrastructure. It was a watershed moment. I learned a lot of lessons in 2007.
The main one was that there is no substitute for informal and rapid information exchange - internally and with our allies. That is how Estonia got through its cyber-crisis - thanks to the support of others, including the United States. Realistically, no country can succeed alone.
Now, I am using that experience at the European Commission in Brussels, where I am responsible for creating a Digital Single Market for the European Union.
In 2007, cyber-attacks were used as a weapon to achieve political goals. This was a strong signal to the whole world that cyber space would very likely be used in the future to attack independent countries. Since then, we have seen that many other countries have been on the receiving end.
Malicious cyber-activity has proliferated. It has become more brazen and sophisticated, more imaginative and international.
Misinformation is another widely used tool of political influence.
In Russia, for example, military doctrine sees cyber operations as part of its tactics for information warfare. Deception, false data, and destabilising propaganda are deemed legitimate tools to convince people to 'buy' the disinformation message as credible information.
It is a sad truth - but if you repeat false information often enough, sooner or later some people start to accept it as true.
Three years ago, EU leaders decided they had had enough - and so a special team was set up to improve Europe's forecasting and response to pro-Kremlin 'information weaponising'.
In its first two years, it identified more than 3,500 examples that contradicted publicly available facts. Here are some from the last month that name the United States directly:
Take this one:
"Americans poisoned Russian ex-spy Skripal and daughter to spread Russophobia". That came from the "60 minutes" programme broadcast by the Russia 24 channel. It is owned by the Russian state.
Or this one, from NTV - controlled by Gazprom Media: "The US intelligence services are planning to kill one of the presidential candidates in Russia".
Both examples push a conspiracy theory, but with no evidence to back it up. And it continues today, multiplied across many languages and repeated daily.
A huge propaganda machine. Who would be in a position to pay for that, if not a government?
In Europe, as in the United States, we remain on the frontlines of these assaults on democracy, threatening to undermine institutions.
A welcome development in this cyber-gloom is that there is now more willingness to name perpetrators, even if it concerns a specific country rather than individuals.
Given the scale and scope of the threats, people should name names, if they can. Collective attribution makes us stronger against the threat, wherever it is from.
The world faces a new strategic environment: one where we should help each other even more.
To me, it is why the EU-US partnership needs to stay strong: for security and prosperity of both sides of the Atlantic.
On cybersecurity, Europe is already working with the United States.
I would like to see more cooperation - perhaps to explore the idea of a secure transatlantic cyber area to deter cyber-attackers.
For example, we have proposed EU certification for cybersecurity products and services. This should be a good basis to discuss and make sure that our cyber standards are aligned on both sides of the Atlantic.
If both sides could agree on common security standards for the IoT, this would set a global standard. Exchanging detailed information about cyber incidents will help to prevent future attacks. We are in the same boat here: if Europe is the target today, the United States could easily be under attack tomorrow.
Transatlantic cooperation on cybersecurity will help to maintain secure and open data flows between the United States and Europe: they are the world's highest, after all. But that also depends on trust.
Trust is easy to break - and difficult to rebuild, as the world has seen with the recent case of Cambridge Analytica.
If people feel their privacy has been violated, their digital profiles misused, then they will react accordingly. And when online trust is eroded, any digital economy will find it hard to advance.
As you probably know, the EU's GDPR enters into full force next month:
-it will reinforce all our data protection rules, give people more control of their data, and set the rules on profiling and data portability.
-it will force companies to be more responsible and accountable in how they deal with our data.
-and it is practical: companies will no longer have to follow 28 different country laws: just one for all of Europe.
While it contains financial penalties for non-compliance, this is a last resort - we have to make sure that the rules work in practice. They are fit to tackle the constant hunger for increased monetisation of data and targeted advertising. They have already changed how businesses deal with our data, and not only in Europe.
Frankly, I would advise companies to invest in data privacy in the first place. Both the EU and United States have a strong commitment in this area.
Here, I am thinking of the Privacy Shield. This makes sure that privacy is respected when the data of Europeans is sent to the United States. It is already being used by more than 2,700 companies; incidentally, also by Cambridge Analytica.
This just shows how crucial it is for the Privacy Shield to be an effective and enforceable instrument. There is a lot still to be done, certainly in the United States.
We are also closely following the FTC's investigation into the Cambridge Analytica case.
Artificial intelligence (AI) is another example. This technology is already with us - globally - and evolving fast. It raises issues of cybersecurity, privacy, ethics as well as protection of fundamental liberties.
These issues concern us all, which is why we should lay down some common principles on how AI is developed and deployed. And we need to do this now.
If we fail to do so, if the West fails to unify - we risk being exploited by those who would use cyberspace as a weapon to harm our free and open societies and economies.
By not acting, we make ourselves an easy target.
Digital innovation should not be abused like this, I think you would agree. The perpetrators are operating on a global and daily basis, and that is not about to change. We have to fight it together - to prevent, deter and respond. Constant vigilance and cooperation. But it is not only about defeating threats. There is a lot to be positive about, especially when we work together.
After all, this is about building a thriving transatlantic digital economy, so that we all benefit. A bright global digital future.