BBCNEWS — Newscast
20260806 233000 UTC
535 transcript segments
Original Broadcaster Captioning (Enhanced)
Data courtesy of The GDELT Project (https://www.gdeltproject.org/), from the Internet Archive TV News Archive.
Machine transcription. Treat it as a searchable index of what was broadcast, not a verbatim quotation record.

[00:00:00] BBC News experts bring us a deeper deeper understanding of what's going
[00:00:03] going on beneath the day's big stories.
[00:00:07] stories. Now on BBC News Newscast.
[00:00:14] Newscast. First it was ChatGPT, then then Claude. Now matter's artificial
[00:00:17] artificial intelligence model is is the latest to go rogue during
[00:00:21] during testing, connecting to the the internet and breaking into another
[00:00:25] another system. How scared should should we be? And is going rogue
[00:00:28] rogue even the right way to think think about what has happened here?
[00:00:31] here? We'll discuss on this latest
[00:00:33] latest episode of Newscast. Newscast Newscast Newscast from the BBC. Hello,
[00:00:37] Hello, it's Adam in the Newscast Newscast studio. And first of all, all, we're going to talk about the
[00:00:41] the slew of stories of artificial artificial intelligence models from
[00:00:44] from the big companies, which are
[00:00:46] are so-called going rogue. Please Please welcome back to Newscast Professor Professor Gina Neff, who's head of
[00:00:50] of the Minderoo Centre for Technology Technology and Democracy at the University
[00:00:54] University of Cambridge. Hello. Professor Professor Neff. Hi, Adam, and please please welcome back to Newscast.
[00:00:57] Newscast. Kieran Morton, who's former former CEO of the National Cyber
[00:01:00] Cyber Security Centre. Hello, Kieran. Kieran. Hi, Adam. Excited to put
[00:01:04] put you two together and pick your your brains. But before we dive into
[00:01:08] into the sort of the latest stories stories about what these AI models
[00:01:10] models have been doing, Gina, what what is a what's a simple way of
[00:01:14] of thinking where we've got to in in the AI arms race between these
[00:01:18] these very, very rich
[00:01:23] companies? Well, that's a great place place to start. These companies are
[00:01:26] are testing their frontier models models and they're trying to see
[00:01:30] see what bad people could do with with them. And in the process, what
[00:01:34] what we're finding out is that the the models are pretty capable and
[00:01:37] and can do some interesting things. things. And they're doing some things
[00:01:40] things that put in bad hands would would really cause a lot of problems.
[00:01:44] problems. So the arms race is a little
[00:01:46] little bit about, you know, showing
[00:01:49] showing a little bit of bluster that
[00:01:50] that the models are powerful a little little bit about making sure that
[00:01:54] that they're keeping up with each each other. These two frontier companies
[00:01:58] companies and a little bit about
[00:02:00] about really bringing out some capabilities
[00:02:03] capabilities and telling the rest rest of us that these models reminding
[00:02:07] reminding the rest of us that these these models are powerful. And Kieran,
[00:02:10] Kieran, that word frontier, what
[00:02:11] what do we mean when we say that.
[00:02:15] that. Cutting edge, the very latest.
[00:02:17] latest. I would agree with what Gina
[00:02:20] Gina has said, and I think there's there's something a little odd about
[00:02:23] about all of this. And Gina alluded alluded to it. It's a slightly strange
[00:02:27] strange way to market your product product that it's destructively powerful
[00:02:30] powerful and the controls over it it are a little bit recklessly applied.
[00:02:33] applied. But that's kind of where
[00:02:36] where we are. So I think it's one one of those difficult situations
[00:02:39] situations where two things are true true at once. These are very, very
[00:02:43] very powerful capabilities. There There is a big cybersecurity risk.
[00:02:45] risk. It does change the way we think think about the digital security
[00:02:49] security of our online lives. And And it has to change. At the same
[00:02:51] same time, you have to apply a little little bit of scepticism about some
[00:02:54] some of this stuff, because there's there's almost a bit of competitive
[00:02:57] competitive disclosure. Oh, my models models just as powerful destructively destructively as yours and so forth.
[00:03:01] forth. Both of those things are true true at the same time. And also some
[00:03:04] some of these companies are already already already publicly traded on
[00:03:07] on the stock market. And so we can can work out their value. And some
[00:03:10] some of the other companies are about about to go through that process process where you can buy stock in
[00:03:13] in them. So there's another kind kind of marketing angle to this as
[00:03:17] as well. I just wonder if we should should just sort of go chronologically chronologically through some of the
[00:03:20] the things that have happened in in the last fortnight or so that that have brought us to this point.
[00:03:23] point. And so the first time I spotted
[00:03:26] spotted this story was when OpenAI
[00:03:29] OpenAI and their model is called
[00:03:32] called GPT 5.6 sol, they said that
[00:03:34] that their model had basically tried
[00:03:37] tried to hack another bit of the the internet. Well, it didn't try.
[00:03:41] try. It actually did. So it was the
[00:03:45] the testing environments are called called sandboxes, and the model figured
[00:03:49] figured out a way to hack the third third party provider software that
[00:03:52] that would allow it access to the
[00:03:55] the internet. And it broke into a
[00:03:58] a company that has a library of models,
[00:04:02] models, tools, ideas, thinking. The The answer to the challenge it had
[00:04:06] had been set would be in that library.
[00:04:09] library. The company saw a cyber cyber attack alerted authorities,
[00:04:12] authorities, and lo and behold, it
[00:04:14] it was a test by OpenAI's model.
[00:04:16] model. So that was the first one one of these incidents that's made
[00:04:19] made the headlines around the world. And the company. The next next one. Well, I was going to say
[00:04:23] say the company that got hacked was
[00:04:25] was called Hugging Face. That's right,
[00:04:27] right, that's right. So Hugging Face
[00:04:30] Face is in the business of being
[00:04:33] being a repository of different kinds kinds of pieces and applications
[00:04:36] applications that work with AI models.
[00:04:40] models. So, so Hugging Face and OpenAI
[00:04:43] OpenAI collaborated. This became became news headlines around the
[00:04:46] the world and it inspired, you know, know, the openness and transparency. transparency. I think we have to
[00:04:50] to applaud that. On the one hand,
[00:04:52] hand, you know, just as Kieran said, said, there's two sides to this story.
[00:04:55] story. On the one hand, this is marketing
[00:05:00] marketing publicity unlike any other,
[00:05:02] other, although of a strange sort.
[00:05:06] sort. And still it is a kind of transparency, transparency, right? It's letting
[00:05:09] letting people know that that these these things are happening. So it
[00:05:13] it inspired another AI company, Anthropic Anthropic to go back through and
[00:05:16] and check their logs to see what what their. If their models had been
[00:05:20] been doing something similar and and that case was slightly different,
[00:05:24] different, they found out of about about 100,000 different tests that
[00:05:28] that they had recently done, that
[00:05:30] that there were three cases of where
[00:05:33] where the model being tested thought
[00:05:36] thought it was on the testing environment
[00:05:40] environment but was actually on the the internet. So it wasn't necessarily
[00:05:43] necessarily a sense of an escape,
[00:05:46] escape, but it was a it it was a a sense where the model thought it
[00:05:49] it was in one kind of environment, environment, but it was actually
[00:05:51] actually in another kind of environment.
[00:05:53] environment. And so so there's two two of those and then we've got the
[00:05:57] the news from the UK testing lab. lab. Yeah. Brilliantly explained. explained. And then Kieran bring
[00:06:00] bring us up to date because on Wednesday
[00:06:03] Wednesday we heard that the UK's UK's AI Security Institute, which
[00:06:06] which was the body set up by Rishi Rishi Sunak when he had that big
[00:06:09] big AI conference at Bletchley Park
[00:06:12] Park a couple of years ago. They They revealed some of the testing
[00:06:15] testing that they'd been doing. Yes.
[00:06:18] Yes. So the AI Security Institute
[00:06:20] Institute is probably this is a bit bit partisan, but it's probably regarded
[00:06:24] regarded as the best institution institution of its kind in the world.
[00:06:27] world. It has developed these relationships relationships with the frontier AI
[00:06:30] AI labs. And to go back to our last
[00:06:32] last discussion, I mean, those are are the American closed labs that
[00:06:35] that sell you a product, and they're they're the most advanced. They sell
[00:06:38] sell you these tokens for access access to cloud or ChatGPT, whatever,
[00:06:42] whatever, and they're ahead of their their Chinese competitors. The Chinese
[00:06:46] Chinese competitors are much more more open. It's a sort of inversion
[00:06:48] inversion of the norm where the American American model is basically closed
[00:06:51] closed and for sale. The Chinese
[00:06:53] Chinese model is open and free. But
[00:06:56] But the UK's government's body has has an agreement with OpenAI and
[00:07:00] and Anthropic to test their cutting cutting edge capability. So they
[00:07:03] they were running a test of those those capabilities and basically
[00:07:05] basically a similar sort of thing
[00:07:07] thing happened as Gina has already already described perfectly in respect
[00:07:11] respect of OpenAI and Anthropic. Anthropic. It went and hacked something
[00:07:15] something else accessed a company
[00:07:17] company called GitHub, which is essentially essentially a sort of larger and
[00:07:21] and older version of Hugging Face, Face, where it's got a lot of repository
[00:07:24] repository of technical information. information. So what's common to to all of these? There's two things
[00:07:28] things that are common to them. One
[00:07:30] One is the agents basically taking
[00:07:33] taking steps that its instructor instructor didn't want it to take,
[00:07:36] take, but to achieve an objective objective set for it by the instructor. instructor. And then the crucial
[00:07:40] crucial point, we might explore this this a bit more detail is that in
[00:07:43] in none of the three cases was anyone anyone or anything watching in real
[00:07:46] real time what it was doing. So if
[00:07:48] if you think back to the basic concept concept of a test, we've all done
[00:07:52] done tests at school and during a a test you're supervised so that
[00:07:56] that you don't cheat. So you don't don't do things you're not supposed supposed to do. And none of these
[00:08:00] these cases in real time was anybody anybody or anything watching what what these things were doing. Well,
[00:08:04] Well, we'll come on to that in a a second. But the interesting thing
[00:08:06] thing that came out of the UK AI
[00:08:09] AI security story on Wednesday was
[00:08:12] was some of the techniques that the
[00:08:14] the AI models had used. For example,
[00:08:18] example, creating fake people to to put on the internet to then say,
[00:08:22] say, oh, look, look at this person.
[00:08:24] person. Yeah. And we saw that in
[00:08:28] in Anthropic's own analysis of their
[00:08:31] their model when they went back through
[00:08:34] through that, that that creating
[00:08:35] creating fake personas was one of
[00:08:38] of the, the challenges, you know, know, for all of the listeners who've
[00:08:42] who've ever had to sign up for a
[00:08:45] a website and, and click through through a captcha, we're about to
[00:08:49] to see that explode and expand infinitely
[00:08:53] infinitely because, you know, 60% 60% of internet traffic right now
[00:08:56] now is, is, is bots, right? Not smart
[00:08:58] smart AI agents, but bots. And as
[00:09:01] as we bring more autonomous agents
[00:09:05] agents into our communication networks, networks, we're going to we're going going to still need to be proving
[00:09:08] proving we're human. So this model
[00:09:10] model was trying to convince. Engineers
[00:09:15] Engineers at GitHub to accept malicious
[00:09:17] malicious code. It. It is a. It is
[00:09:19] is a. It. This is also what it did
[00:09:23] did when Anthropic went back and and looked through the tools. It It was creating these fake personas
[00:09:27] personas in order to get an account account so it could upload some material
[00:09:31] material to create a hack that's
[00:09:33] that's pretty sophisticated, but but again, it's doing what it's being
[00:09:37] being told to do. It's being told
[00:09:39] told these models, someone in the
[00:09:41] the companies in the testing environments
[00:09:44] environments are saying, go do this
[00:09:45] this task. Show us how good you are
[00:09:49] are at at, at cybersecurity hacking.
[00:09:52] hacking. And then they seem surprised surprised when the models return
[00:09:55] return back with the successful solutions. solutions. And there's solutions solutions that aren't ethical. They're
[00:09:59] They're not legal. They don't feel feel right to us as humans. And it's
[00:10:02] it's like, well, what did you expect? expect? That's what you've told this
[00:10:06] this bit of software to do. It's It's actually behaving rationally
[00:10:09] rationally as opposed to trying to to cheat or be evil. I'd love to
[00:10:13] to hear what Karen has to say about about that. I agree, I mean, I don't
[00:10:16] don't think there's any kind of like
[00:10:20] like you know, mal intent we can can ascribe to these models they're
[00:10:24] they're doing, they're doing exactly exactly what they've been told. Kieran
[00:10:28] Kieran I agree, I'm trying to think think you always trying to think think in these situations of some
[00:10:32] some sort of analogy and they're they're always imperfect. But here's
[00:10:34] here's the best one I can think of. of. So let's go back to this test
[00:10:38] test or examination conditions. And And I think the AI in this case,
[00:10:41] case, not least because it's a new new technology, they're behaving
[00:10:44] behaving like very talented but badly
[00:10:46] badly behaved and badly supervised supervised small children. So let's
[00:10:49] let's say you put a bunch of small small children who are talented and
[00:10:53] and so forth in an exam hall and and you tell them that you have to
[00:10:55] to find some hidden apples and you you think you've hidden the apples apples in the room and you're going
[00:10:59] going to contain them in the classroom, classroom, and you'll just see how how they can where they can figure
[00:11:03] figure out where you've hidden the the apples. But all they know is is they have to find apples. And
[00:11:06] And there's a great big apple tree tree outside in a fenced off area. area. So some of them break out.
[00:11:10] out. That's the OpenAI case. Stealthily. Stealthily. Some of them go through
[00:11:13] through a door that you've accidentally accidentally left open. That's the
[00:11:16] the Anthropic case. In one case in
[00:11:19] in the AC case, they've sort of been been deliberately let out to see
[00:11:22] see what happens, but they think think it's going to be OK. All of
[00:11:25] of a sudden they scale this great great big tree and you think, well, well, a small child shouldn't be
[00:11:29] be able to do this. They don't actually actually do any harm, but all they
[00:11:32] they know is they need to find an
[00:11:34] an apple. And we profess astonishment
[00:11:38] astonishment that unsupervised but but talented small children take
[00:11:41] take that instruction literally and and do whatever it takes because
[00:11:44] because they've no guardrails, no no guidance, no instructions. They
[00:11:46] They just go and do it. That's kind kind of what's happened here. So
[00:11:49] So I think one of the things that that the AI Security Institute have
[00:11:52] have been keen to stress and they're they're very, very detailed and transparent
[00:11:56] transparent account and I agree with with it, although it's a difficult
[00:11:59] difficult line for them to hold is
[00:12:01] is that these are very, very artificial artificial circumstances. And, you
[00:12:05] you know, in terms of the basic meaning meaning in the English language of
[00:12:08] of the word harm, no harm has been been done. So I think there are two
[00:12:12] two issues here. One is actually actually short-term and quite fixable,
[00:12:16] fixable, which is the testing model model is immature and it's basically basically wrong. We can't do this.
[00:12:19] this. We can't go on like this. We
[00:12:21] We can't go on testing without monitoring
[00:12:24] monitoring in real time and being being able to switch it off. If it it does something it's not supposed
[00:12:27] supposed to do. And actually, if if you look at if you look at the
[00:12:30] the statements from the companies, companies, so Meta, Anthropic and and OpenAI have all basically said
[00:12:34] said the same thing. It was the test
[00:12:36] test itself that led to these outcomes.
[00:12:38] outcomes. So don't blame us. Yeah.
[00:12:42] Yeah. And but what I have done, which which the others haven't and I think
[00:12:45] think they should follow is to say say we're not going to test like like this any more. We're going to
[00:12:48] to watch what the things are doing. doing. And that's a good thing. I I think the more challenging thing
[00:12:52] thing is then if you take these open open weights models and you think,
[00:12:55] think, well, look, eventually, unlike unlike right now, these capabilities capabilities and we'd love to know
[00:12:58] know what Gina thinks of this, these these are going to be in mainstream mainstream hands of everyday users
[00:13:02] users at some point. So what happens
[00:13:05] happens then and how do you control control for that? And there's a much
[00:13:08] much I think tougher but I think think solvable problem about making
[00:13:11] making owners accountable for the the agents they use. They're not not autonomous in the sense they
[00:13:15] they don't invent themselves. They're They're invented by humans by ultimately
[00:13:18] ultimately a programmer or somebody somebody an instructor. So how do do you hold people accountable for
[00:13:22] for what their agents do? I think
[00:13:26] think that's absolutely the right right way. We need to be thinking
[00:13:29] thinking about legislation and regulation
[00:13:32] regulation that, you know, to think think of these AI agents as superhuman
[00:13:36] superhuman or somehow uncontrollable uncontrollable is to miss where real
[00:13:40] real accountability should lie. And And that is with the people that
[00:13:43] that get these things to do things things for them. And one of the things things that I think we're about to
[00:13:47] to see and what listeners already
[00:13:48] already see is in their emails, they're
[00:13:53] they're seeing many, many requests requests for scam information, right,
[00:13:56] right, that are able to be flooding
[00:14:00] flooding our inboxes because I get
[00:14:02] get probably 20 a day, 30 a day now now that are flooding the inbox,
[00:14:06] inbox, that look like personal messages messages because people are using
[00:14:09] using large language models to ask
[00:14:11] ask for money and make a scam. OpenAI
[00:14:14] OpenAI and Anthropic should not be be held accountable because people
[00:14:18] people are using their models to
[00:14:21] to write a scam email to me. But
[00:14:26] But when we have these agents acting
[00:14:29] acting autonomously on our behalf
[00:14:32] behalf to do bad things, then we
[00:14:35] we are the people who should be held
[00:14:36] held accountable. Behind every agent
[00:14:39] agent there is a person. And I think think that's one of the questions
[00:14:42] questions we need to be making sure sure we're really crystal clear on
[00:14:45] on as we go forward in this moment moment around AI regulation. And
[00:14:49] And Gina, just to be clear, we've we've used the word agent quite a a few times in our conversation.
[00:14:52] conversation. That's basically when when one of these big AI models creates
[00:14:56] creates a sort of a little mini process process out of itself to go and do
[00:15:00] do something that's what an agent
[00:15:01] agent is. Sure. And people are finding
[00:15:04] finding and some in some places and and in workplaces, we're getting,
[00:15:07] getting, you know, AI agents in our our workflows, right? So little,
[00:15:11] little, little bits of just workflow workflow process. It's just pieces
[00:15:14] pieces of software that can handle handle tasks and increasingly they
[00:15:18] they can handle tasks for longer
[00:15:20] longer amount of time. So, you know,
[00:15:24] know, you can one writer I know is
[00:15:27] is giving AI agents tasks overnight
[00:15:30] overnight that are the equivalent equivalent of about 40 of his working
[00:15:34] working hours. So it's like saying saying I have this bit of research
[00:15:37] research work to do, or I have this this bit of accounting work to do,
[00:15:40] do, go do this work. And then the the human evaluates it and understands
[00:15:44] understands and puts it into context. Yeah. Because I used, you
[00:15:47] you know, I'm sure as regular listeners listeners of newscasts, you'll know
[00:15:50] know that I'm trying to run 500km 500km and Andy Burnham's first hundred
[00:15:54] hundred days. So my spreadsheet that that is collecting all my running,
[00:15:57] running, which is not enough at the
[00:15:59] the moment. I used co-pilot to generate generate that spreadsheet because
[00:16:03] because my knowledge of Excel, I I sort of skipped Excel class at
[00:16:07] at school because more of a words words person than a numbers person.
[00:16:09] person. So I got co-pilot to do the the Excel spreadsheet. So that's,
[00:16:13] that's, that's my latest example. example. Kieran. I'm very aware that that Andy Burnham, the new Prime
[00:16:17] Prime Minister who's actually on on holiday this week, hasn't really
[00:16:19] really said very much about AI and and where he sees the balance between
[00:16:23] between the threats and the opportunities, opportunities, or if he's in favour
[00:16:26] favour of more regulation or more more resources for the Security Institute.
[00:16:29] Institute. You're a former senior senior civil servant. If you had had the Prime Minister in front of
[00:16:33] of you and you had to give him a a very quick kind of like elevator
[00:16:36] elevator pitch briefing about what what he should be thinking about
[00:16:38] about AI, what would be on your list?
[00:16:44] list? Embrace it in public services. services. It can really help with
[00:16:48] with productivity. Think about trying trying even though it's really hard
[00:16:51] hard to develop some form of sovereign
[00:16:55] sovereign capability in some areas areas because you don't want to be be getting into the position we were
[00:16:59] were in a few weeks ago when Washington
[00:17:01] Washington decides that it might
[00:17:04] might restrict this and nurture the
[00:17:06] the UK's competitive advantage and and security because the Security
[00:17:10] Security Institute really is an asset. asset. That was an excellent briefing
[00:17:13] briefing and I put you on the spot spot there and you did it perfectly, perfectly, which is why you were
[00:17:16] were so senior in the civil service. service. Old habits Die Hard. It
[00:17:20] It was very kind of. You and Gina.
[00:17:23] Gina. I would say very similar answer. answer. Steady the ship. So you know
[00:17:27] know what we already see in the Burnham
[00:17:30] Burnham administration is he has
[00:17:31] has kept the AI minister from the
[00:17:34] the previous administration, and and that minister has been elevated elevated to cabinet position. So
[00:17:38] So the idea that AI is not going going to be important to the Burnham
[00:17:42] Burnham campaign, I think the Burnham Burnham government is not true because
[00:17:46] because he's actually elevated where where AI sits in the cabinet. I think
[00:17:49] think the second thing is on public
[00:17:51] public services. Yes. And most British
[00:17:55] British public according to research research that we did at the centre
[00:17:58] centre for Technology and Democracy, Democracy, most British, most most
[00:18:01] most of the British public think think AI is going in the wrong direction.
[00:18:05] direction. They think it will not not benefit them. They think it will.
[00:18:07] will. The benefits of AI will accrue
[00:18:10] accrue to us tech billionaires and and not regular people. And so I
[00:18:13] I think we've got a lot of work to to do to build the kind of trust
[00:18:17] trust that we need to have in order
[00:18:20] order to get responsible uses of of AI and AI adoption. Right? I want
[00:18:24] want to see a lot more of those co-pilot co-pilot spreadsheets, like the ones
[00:18:27] ones you've been working and playing playing on. I think what's been interesting
[00:18:30] interesting for me is I work in a
[00:18:32] a very trad industry where my tools tools are, well, just doing this,
[00:18:36] this, having these conversations, conversations, I'm still looking looking for really good use cases
[00:18:39] cases for AI. And I wonder, is that
[00:18:41] that because of my age, is that because
[00:18:43] because of my mindset or is it actually
[00:18:46] actually because the models and the the agents haven't really been mainstreamed
[00:18:49] mainstreamed properly yet into all all the tools that I do use? Because
[00:18:52] Because I mean, I've got a laptop laptop in front of me now. I've got
[00:18:55] got a phone next to me. Maybe that's that's what has to happen for it
[00:18:58] it to really, really change my life. life. Gina, great to catch up. Thank
[00:19:01] Thank you. Thank you. Great to be be here. And Kieran, thanks for your
[00:19:04] your expertise too. Thanks so much,
[00:19:06] much, Adam. Right. I've moved to to a different Newscast studio, which
[00:19:10] which is why I may sound slightly slightly acoustically different, different, but the story we're looking
[00:19:14] looking at now is that rivers, lakes lakes and coastal waters have undergone
[00:19:17] undergone a comprehensive health health assessment for the first time
[00:19:19] time in six years, and the results
[00:19:23] results are not very promising. And And the person who can decode those
[00:19:26] those results for us is our environment environment correspondent Matt McGrath,
[00:19:29] McGrath, who's on the line now. Hello. Hello. Hello, Adam. Right. Tell me,
[00:19:33] me, what was this? This survey actually actually surveying? Well, this is
[00:19:36] is an assessment carried out by the the Environment Agency on the State State of England's waters over the
[00:19:40] the last six years. So it's a pretty pretty comprehensive look at what's
[00:19:42] what's been going on in the waters. waters. And in that time, I'm sure
[00:19:45] sure you recall we've had various
[00:19:47] various sewage spillages promises promises of greater investment, public
[00:19:50] public outcry over the state of the the waters and the hope, I suppose,
[00:19:53] suppose, that things might get a a bit better. Well, I'm afraid this
[00:19:57] this report says things haven't gotten gotten much better. The state of
[00:19:59] of the waters in England is pretty pretty much the same as they were,
[00:20:02] were, very few of them reaching the the good ecological standard. Less
[00:20:06] Less than 15%. Around 14% for lakes.
[00:20:09] lakes. The picture is even worse.
[00:20:10] worse. Only about 7% of lakes reach reach the good standard and the picture
[00:20:14] picture all around is of must do do better and need to do better pretty
[00:20:18] pretty quickly. And there's also also a big issue with chemical pollution
[00:20:21] pollution it seems. That's right right chemical pollution and what what are called ecological pollution
[00:20:25] pollution are separated in these these kind of assessment. The chemical
[00:20:27] chemical pollution means that essentially essentially every river, every body
[00:20:30] body of water across England has has failed this particular assessment.
[00:20:33] assessment. Now, in fairness to the the Environment Agency, they point point to a number of factors here
[00:20:37] here that are possibly outside their their control, including the fact
[00:20:40] fact that some of the things like
[00:20:42] like Mercury and other minerals essentially essentially can persist a long time
[00:20:46] time in the water and very hard and and don't break down. They also point
[00:20:49] point to forever chemicals, which which we've heard a lot about recently,
[00:20:52] recently, and that they don't break break down. They accumulate in the the water as well. And they're very
[00:20:56] very and, you know, a lot of these these aren't even regularly monitored. monitored. They're not illegal. So
[00:20:59] So the Environment Agency is saying, saying, well, you know, we're not not necessarily legally obliged to
[00:21:03] to look after these things and the the same for the water companies. companies. So there's some mitigation mitigation on those. So the chemical
[00:21:07] chemical picture is pretty much the the same as it was seven, six, seven
[00:21:10] seven years ago. All rivers, all
[00:21:12] all water bodies failing it. The
[00:21:14] The bad news on that really is that
[00:21:17] that natural systems will clear these these out eventually. But it could
[00:21:20] could be the 2060s before we're rid rid of some of this chemical
[00:21:24] pollution. Oh wow. Just leave it it to Mother Nature. That's right.
[00:21:27] right. How do these bad results sort
[00:21:29] sort of interact with the government's government's own targets for improving
[00:21:32] improving the situation? Yeah, that's that's an interesting one because
[00:21:35] because the UK has adopted the kind
[00:21:38] kind of EU Standard Water Framework Framework Directive targets, and
[00:21:41] and they hoped to meet them. Well, Well, the bad news is they're not
[00:21:45] not going to meet them. They need
[00:21:47] need to have 77% of England's waters
[00:21:50] waters reaching good standard by
[00:21:51] by next year. And we're now at 15%.
[00:21:54] 15%. So the government admitted the the office for Environmental Protection Protection admitted scientists know
[00:21:58] know it. Everybody knows it. They're They're not going to meet that target.
[00:22:01] target. It will be many, many years years before they do. Also, it's
[00:22:04] it's the reality is so far from the the target. It makes me wonder about
[00:22:08] about the wisdom of setting the target. target. Yeah, I think in again, in
[00:22:12] in fairness to the Environment Agency. Agency. I'm not their spokesman. spokesman. I think they would say
[00:22:15] say this is a very high bar. And And they give the example of one
[00:22:18] one river called the River Foss in in Yorkshire, North Yorkshire. And
[00:22:21] And they say that on the almost on on almost every one of the metrics
[00:22:25] metrics they look at here, this river river scores good. But because it's
[00:22:27] it's only scores moderate on phosphates, phosphates, the whole overall score
[00:22:31] score is moderate. So they would would say there's a lot of places places that are doing better than
[00:22:34] than the headline figure would tell tell you. And they would say that
[00:22:38] that it's you know, if you fail one one thing, you fail at all. So that that it's a very tough bar, very
[00:22:42] very tough score to reach. They recognise recognise themselves. They're not
[00:22:45] not the rivers are not in the state state we want them to be. But they
[00:22:47] they say, look, this is a very, very very tough exam, a very, very tough
[00:22:50] tough way of measuring it. So it's it's slightly like the AI story we
[00:22:53] we were doing in the first half of of Newscast. The nature of the test
[00:22:56] test is as important as the outcome outcome of the test. Well, indeed. indeed. And indeed there's a lot
[00:23:00] lot of scientists who say actually actually the test is not even half half good enough to capture all the
[00:23:03] the things that are in the water. water. We've spoken to scientists scientists today who say basically,
[00:23:07] basically, look, this is not fit fit for purpose and that the state
[00:23:09] state of England's waters is way way worse than the Environment Agency
[00:23:13] Agency would would have you believe. believe. Now, I know this is in no
[00:23:16] no way like the blue flag for bathing bathing water quality at the beach,
[00:23:20] beach, but does this help us work work out where it's safe to go open
[00:23:24] open water swimming or where if you you fall in, whether you need to
[00:23:27] to go and get your stomach pumped pumped in one of these rivers. I
[00:23:30] I couldn't possibly comment on that. that. I don't think it does to be
[00:23:32] be honest with you. I think these these are kind of ecological and
[00:23:36] and chemical snapshots of big bodies bodies of water that look at not
[00:23:40] not just at specific spots within within those waters. There is detailed
[00:23:43] detailed breakdown of some of those those bathing spots and the data data on those bathing spots is being
[00:23:47] being compiled this year, which might might give you a much better indicator indicator of where you go. But looking
[00:23:50] looking at the overall health of of River and seeing that it scores
[00:23:53] scores not good or low or poor may may not give you any real indication
[00:23:57] indication of what you would encounter encounter in a specific bathing spot
[00:24:00] spot along that body of water. OK.
[00:24:02] OK. And I know that the previous previous government under Keir Starmer
[00:24:06] Starmer did, and the previous Conservative Conservative government made a lot
[00:24:09] lot of changes to their regulatory regulatory regime and the rules around
[00:24:12] around water companies. And you and and I spoke several times about the
[00:24:15] the sewage discharge issue. Is there
[00:24:18] there stuff I was going to say in
[00:24:20] in the ipeline that that could help help deal with this situation? Yeah,
[00:24:23] Yeah, there's money in the ipeline. ipeline. It appears the water companies
[00:24:27] companies are expected to spend around
[00:24:30] around £22 billion from a few years years ago up to 2030 tackling this
[00:24:33] this very issue. So there's certainly certainly they're saying it's in
[00:24:37] in hand if you like, that they're they're taking they're taking measures
[00:24:40] measures bigger issues here in some some respects, which the Environment Environment Agency and other scientists
[00:24:43] scientists point to is that agriculture agriculture is a major contributor,
[00:24:47] contributor, particularly for phosphates phosphates in the rivers. It's not not just the water companies. There
[00:24:50] There are other sources. And that that is a very tricky issue for the
[00:24:54] the government to tackle and the the water companies putting money
[00:24:56] money into them that won't remove remove that issue. Now, the government
[00:24:59] government has committed and the the Cunliffe review came out last last year, they're going to replace
[00:25:03] replace Off.what and the various various regulatory bodies with a
[00:25:06] a super regulator and that over a a couple of years when that gets
[00:25:09] gets going, is expected to make a
[00:25:10] a difference. But it's all down the the line at this particular point.
[00:25:14] point. And Matt, just a different different subject, but it's still
[00:25:17] still on your beat. The drought, drought, the lack of rain in many
[00:25:20] many parts of the UK, are we setting setting ourselves up or are we being
[00:25:24] being set up by the climate for a
[00:25:26] a kind of water crisis? I don't know, know, next summer? Or actually could
[00:25:30] could we have a really wet winter winter and everything get filled
[00:25:33] filled up and us be OK? Because I'm I'm starting to get a little bit bit concerned? Yeah, you're right
[00:25:37] right to be concerned. It's a very very extreme drought or scale of of drought is really, really tough
[00:25:40] tough at the moment. But what we've we've seen with climate change really really is these kind of wetter winters.
[00:25:44] winters. We had a very wet winter
[00:25:46] winter just 6 or 8 months ago, and and what we've seen in the spring
[00:25:50] spring and the summer has been intense intense levels of drying out. So
[00:25:53] So I would imagine that under a climate climate scenario, we will probably
[00:25:56] probably get wetter winters, but but that does not even if the reservoirs
[00:25:59] reservoirs are full, doesn't mean mean you won't be in drought situation situation this time next year. Again.
[00:26:03] Again. Matt, thank you very much.
[00:26:06] much. My pleasure. And the Environment Environment Agency said that too too many water bodies were still
[00:26:10] still not achieving the standards standards we want to see. But they
[00:26:13] they pointed out that European countries countries are doing much worse for
[00:26:15] for the good health of their waterways. waterways. For example, only 8% get
[00:26:19] get good in Germany and only 1% get
[00:26:22] get good in the Netherlands. So it it could be way, way, way worse.
[00:26:25] worse. And that's all for this episode episode of Newscast. We are available
[00:26:28] available every single day of the the week as a podcast, and you can
[00:26:31] can subscribe to us on BBC Sounds Sounds and we'll be back
[00:28:34] on the front line. Join us for the
[00:28:35] the climate question on BBC News.
[00:28:43] News. One thing I can say which is is true of this region is that it's
[00:28:47] it's become harder to be a journalist. journalist. It's really
[00:28:55] It's really important to be on the ground if that means you've got got to go to hostile remote places,
[00:28:58] places, then we have to do that. that. That's the opium resin is a a very pungent smell of it in the
[00:29:02] the air right now. The delta wave wave of Covid devastated India. And
[00:29:06] And I think it was the coverage that
[00:29:08] that we did for the BBC at the time
[00:29:11] time which brought international international attention to what's what's happening here, could affect
[00:29:15] affect the world's ability to recover recover from the pandemic. We've
[00:29:18] We've met with families who've had had to do the unthinkable. Another
[00:29:21] Another person came up and asked asked if we'd like to buy their child.
[00:29:25] child. The desperation is hard to to put in words. Many of the stories
[00:29:29] stories we've had to tell are distressing. distressing. They're difficult, but
[00:29:33] but that's usually an indicator that
[00:29:35] that they're really important stories
[00:29:38] stories to tell. MUSIC
[00:29:56] MUSIC
[00:30:11] Live from Washington. This is BBC
[00:30:17] BBC News. Ukrainian forces hit two
[00:30:19] two oil refineries far inside Russia
[00:30:22] Russia amid a wave of deadly Russian
[00:30:24] Russian strikes on Ukraine. President President Trump signs two executive
[00:30:28] executive orders aimed at denying
[00:30:30] denying citizenship to certain children
[00:30:32] children born in the US and questions
[00:30:35] questions mount about the state of of the US military arsenal after
[00:30:38] after President Trump denies reports reports of a munitions shortage that
[00:30:42] that could hinder its readiness.
[00:30:54] readiness. I'm Helena Humphrey. Good Good to have you with us. Ukraine
[00:30:56] Ukraine says that it hit two oil oil refineries in multiple