00:00Thank you, Daniela.
00:02My co-founders and I started Anthropic because we believe that the most powerful technology
00:07of our time should be built for the benefit of humankind.
00:11Daniela and I lost our father to a disease that, just a few years later, became curable.
00:16I started my career in biology, hoping to help future cures come sooner.
00:20In the lab, I saw the vast thicket of complexity that scientists face when they're trying
00:25to understand disease and develop new treatments.
00:28So many of the scientific tools that can cut through that thicket, tools for gene editing,
00:34tools for gene synthesis, tools for controlling brain cells with light, started as pure ideas.
00:40In other words, they came from thinking, from intelligence.
00:44Human intelligence has already made diseases like plague, smallpox, and polio history for
00:50most of the world.
00:51But progress in other diseases has been too slow for so many patients and their families.
00:56Going from a great idea to a treatment can take decades.
01:00It took many years from the first discovery of a strange repeating pattern in DNA to the
01:06approval of the first treatment for a disease using CRISPR.
01:11Nature is full of these patterns, but we've had too little intelligence to decipher most
01:16of them.
01:16Too few scientists with too few time.
01:19Now we can build that intelligence.
01:21Whether it's protein folding, drug design, cancer genetics, or clinical trial administration,
01:27scientists across many areas of biology and medicine are finding Claude can help them do
01:32their research in a fraction of the time it would have taken just a couple of years ago.
01:36I believe sufficiently powerful AI working in concert with humans could compress a century
01:42of scientific and medical progress into a decade.
01:46I don't want to get ahead of myself.
01:48At this point, all we have are early results.
01:51Experiments and clinical trials will still take time.
01:53And although they involve slow and ponderous processes, I believe these processes can be
01:59substantially accelerated by AI.
02:02We hope that in the future, fewer families miss a cure by a few years.
02:07Medicine is just a particularly vivid example, but there are many other applications of Claude.
02:13Our customers, the world's most ambitious companies, nonprofits, and individual people,
02:18are already applying the intelligence we've built to stimulate economic growth, lift people
02:24out of poverty, protect democracy and free societies, even make everyday services like the
02:30DMV work better.
02:31However, we're proud to work alongside them, and I'm extremely optimistic about what we
02:36can do together in the future.
02:38At the same time, we've always been honest that this technology has very real risks.
02:43These range from misuse of the technology by individuals or by nation states, to economic
02:49disruption, to losing control of AI systems.
02:52We discuss these risks not because we think they are inevitable and we are building anyway,
02:57but because we think they are avoidable and we can avoid them.
03:01Avoiding them depends on the choices made by the people building AI.
03:05There is a gigantic difference between how AI will go if we pay little attention to safety
03:11versus how it will go if we are obsessed with safety.
03:14And an obsession with safety and public benefit has been built into Anthropics DNA from the start.
03:21From safeguards against biological and cyber risks to alignment training and testing to looking
03:26inside our models like an x-ray, Anthropics has pioneered methods for understanding and
03:31steering the behavior of AI and keeping it under human control.
03:35There is much work still to be done, and the most difficult safety work lies ahead of us.
03:40The incredible talent of the company that has enabled us to build the most capable AI systems
03:46will need to be directed, more so than it has already, towards double and triple checking
03:51that those AI systems are safe and remain under human control.
03:55And to state it unequivocally, we will pace the training and deployment of AI systems,
04:00taking as long as is necessary to have confidence that they are safe and will remain under human control.
04:06We will not bend to either business nor political pressure.
04:10Today puts the company in a stronger financial position and in doing so only strengthens our
04:15resolve to do the right thing.
04:17If we can show it's possible to make the right choices on safety and still build an incredibly
04:22successful company, others in our industry will be inspired to do the same.
04:27We call this a race to the top.
04:29When the race to the top is working properly, it doesn't matter which company has the best
04:34models, everyone wins because everyone is building safely.
04:38The race to the top works best when driven from the front of the pack and we fully hope and
04:42expect to do so.
04:43But the goal is to make the whole industry better, regardless of whether Anthropic is
04:48winning or losing at any given time.
04:50The advent of machine intelligence is one of the seminal moments in human history and we
04:54have a rare opportunity to shape it for the better.