N1BOS — broadcast
20261012 090000 UTC
365 transcript segments
Google Cloud Speech-to-Text API (Chirp) + Gemini 2.5 Flash Non-Thinking
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] constructive, we tried to certainly answer some
[00:00:03] questions that bother us, that is, follow us from last
[00:00:07] week, our reporters in the continuation of the program certainly
[00:00:10] follow similar topics, stay with N1 and
[00:00:14] Newsroom which starts in an hour, through which you are led by colleague
[00:00:17] Nerma Turić, Savatić, with, of course, as I said, the most important reports
[00:00:21] of our reporters. Thank you once again for
[00:00:24] and factories? or a your attention.
[00:00:32] plus, more and more of
[00:00:43] You watched a new day friends of Violeta's show.
[00:01:25] Every other day we have to bed up the castle with fresh
[00:01:28] straw. so that's done with
[00:01:31] diesel. far from the forest, the rising cost of
[00:01:35] diesel is also being felt in this field in North
[00:01:38] Wilture. much of the food we consume
[00:01:42] depends on diesel, the beef at this farm goes to
[00:01:45] MNS, but the cost of producing it is skyrocketing.
[00:01:49] it's terrifying actually,
[00:01:52] with a special formula based because you're looking at the saying, it's becoming unaffordable, but will the
[00:01:55] diesel stocks dry up? what happens if they dry up? we've only got a
[00:01:58] certain amount of diesel. we can store here at the farm, how much can you stock
[00:02:02] pile? how long will this go on for? without
[00:02:05] diesel we we really wouldn't be able to to
[00:02:09] feed our cattle, let alone feed the
[00:02:11] people,
[00:02:14] and before farm even reaches fork, the beef Liz
[00:02:18] books. A large share of produces has got to get there.
[00:02:21] aage firms are some of the biggest consumers of diesel with
[00:02:25] costs nearly doubling since the start of the year. it's really
[00:02:29] hard because you get... people ringing up and they say, oh, can you give me a
[00:02:32] coat to um, take a
[00:02:35] delivery from one place to another? you ask the details and things like
[00:02:39] that, and they say, how much? you try and pass
[00:02:42] to the customer just a little bit, but you don't want to end up losing the
[00:02:45] custom. this is just literally like the
[00:02:48] last nail in the coffin really of everything else that a
[00:02:52] even more. Forget about haulish company's got to deal with, and now we're having to deal with the price of the
[00:02:56] fuel. exclusive analysis given to sky news shows
[00:02:59] that the... raising the cost of diesel is adding 35 million
[00:03:03] pounds a week to Hawlier's fuel bills in the UK.
[00:03:07] The cost of running a long distance Lorry has gone up by more than
[00:03:10] £9,000 year and data shows electric trucks
[00:03:14] are still rare. 97% of new
[00:03:17] lorries registered last year were
[00:03:19] diesel. All this ultimately means the
[00:03:23] price we pay at the supermarket goes
[00:03:26] up. So how long will all this go on for? Well, the 100 million.
[00:03:30] extra barrels of oil and diesel announced on Friday is welcome
[00:03:33] news, but there are questions over how long it will take to have an impact on the
[00:03:37] forect and also how long those fresh reserves
[00:03:41] will actually last? so this release, which
[00:03:44] is approximately about 17% of
[00:03:48] of of the overall reserves which is located mainly in France and
[00:03:51] Germany um would probably buy you time for
[00:03:55] 10 days and assuming that the entire reserves are being
[00:03:58] released, it's a temporary measure, it will have some
[00:04:02] impact on prices, but it does not address the
[00:04:05] fundamental challenges, which is um, the
[00:04:08] shortage is coming from two live conflicts,
[00:04:12] one in Europe and the other one in the middle
[00:04:14] east.
[00:04:16] those conflicts have led to the geopolitical tussel over
[00:04:20] diesel this week. the impact on us and the
[00:04:23] supply chains we rely on could still be
[00:04:26] felt for some time to come. dam.
[00:04:41] Broken villages, burnt cars on the roadside and damaged
[00:04:45] anti-drone nets, we are approaching the artillery positions, we set off
[00:04:48] in the dark to avoid becoming a target for Russian
[00:04:51] drones.
[00:04:56] In the dugouts, the boys have a large RMRL Czech Vampire hidden.
[00:05:00] It is from this that the soldiers have been destroying the occupiers around
[00:05:03] Avdiivka for many years. We could afford to drive a kilometer from
[00:05:07] the front with 40 pieces and calmly
[00:05:10] leave. My crew worked four packages in one day,
[00:05:14] and that was a small record for us, but
[00:05:18] today, due to the abundance of drones, it is too risky to drive a large vehicle,
[00:05:21] it immediately becomes a target for the enemy, so the Vampire operates
[00:05:25] exclusively during Russian assaults. when columns are moving,
[00:05:28] mechanized assaults are happening, when there are questions,
[00:05:32] for example, about the accumulation of infantry,
[00:05:35] there is data that in this square there is a command
[00:05:39] center, ammunition depots,
[00:05:44] accumulations, for example, drone operators arrive in a pile,
[00:05:48] and then disperse. Instead, for
[00:05:51] daily work, the partisan is increasingly used. The Bandera-mobile, also known as the Trident,
[00:05:54] say the fighters. Three
[00:05:57] pipes, a trident.
[00:06:01] By the way, patriotically and quite
[00:06:04] this machine. Many times smaller than a standard multiple
[00:06:07] rocket launcher, it also does not go out into open terrain, as
[00:06:11] before, hidden. It fires from the forest belt, the guys have been
[00:06:14] working with the partisan for a little over six months, they are satisfied
[00:06:17] with its work, this is finding UAV
[00:06:20] crews, which we work on antennas, on the type of
[00:06:23] dugouts, well, and some small vehicles
[00:06:27] that are hidden in
[00:06:30] plantations. Each rocket weighs over 60
[00:06:33] kg, despite such weight, they load the partisan
[00:06:36] in a matter of minutes. This Res weighed plus-minus 62
[00:06:39] kg, so this is a little bit under...
[00:06:42] artillery is the life of our infantry. Gunner Leshy, a former
[00:06:46] infantryman, knows firsthand how even three shots
[00:06:49] can radically change
[00:06:53] they're constantly covering you, and
[00:06:56] 24/7 our side is working, no matter
[00:07:00] what artillery, well, in our
[00:07:03] sector even S60, when it worked,
[00:07:07] they fell silent, it's a very ancient system,
[00:07:11] it's anti-aircraft. The partisan's task this morning is not
[00:07:14] to destroy, but to cover. He will work on the positions of enemy
[00:07:18] drones. During this time, our fighters will be able to
[00:07:21] rotate at the front line. Our guys go out to positions, and we
[00:07:25] work on
[00:07:28] the take-off
[00:07:31] FPVs. Clarification of coordinates,
[00:07:35] adjustments and pencils. This is what the defenders call
[00:07:38] the rockets from Partisan, they fly at the occupiers.
[00:07:45] Aerial reconnaissance reports, three shots hit the target area,
[00:07:48] which means that the Russian position and equipment are damaged, while
[00:07:52] the invaders were hiding, our soldiers were able to break through
[00:07:55] to their positions. Darya Nazarova, Andriy Khupovka, Zaporizhzhia
[00:07:59] TSN, 1+1, marathon, united news.
[00:10:13] Robots are sprinting down racetracks in China. One of the
[00:10:17] world's biggest AI companies is cutting apart
[00:10:20] books by the thousands, and Mark
[00:10:22] Zuckerberg says the future belongs to... personal superintelligence
[00:10:26] from humanoid robots to AI training data and silicon
[00:10:30] valley's latest promises. what do these stories mean for
[00:10:33] users? let's get into it.
[00:10:36] robots that perform martial arts, robots that play
[00:10:40] soccer, and now robot that allegedly beat usain bolt's
[00:10:44] 100 meter record. is this a
[00:10:46] glimpse of what's coming to our homes
[00:10:49] clever PR sent? Because that's what some experts call the world
[00:10:53] humanoid robot games, a... massive PR campaign for the
[00:10:56] Chinese robotics industry: will physical AI really change our
[00:11:00] lives and what does China's push for humanoid robots mean for
[00:11:04] you? Take this guy for example: the
[00:11:08] robot covered 100 m and 9.39 seconds
[00:11:11] and later improved to 8.86.
[00:11:14] Is this a marketing ploy or a technological
[00:11:17] breakthrough? In short, it's both. To move at the
[00:11:21] speed, the robot needs the right materials, sensors and
[00:11:24] software. that can help it adjust its balance in
[00:11:26] milliseconds. making a two-liked machine move fast
[00:11:30] without damaging its joints is a challenge. and by the way, the
[00:11:34] crash at the end of the race isn't necessarily a bad thing for
[00:11:37] developers. some of the things that we humans
[00:11:40] think super easy, like walking, stopping,
[00:11:44] just standing is very complex
[00:11:47] task. testing these under these conditions, um,
[00:11:51] every failure is data. running fast
[00:11:54] shows. mechanical capability, but it's not a
[00:11:57] matter of autonomous decision making. this was
[00:12:00] required at other, less flashy, but more demanding
[00:12:03] competitions at the world humanoid robot games. robots were given
[00:12:07] problem-solving tasks that required autonomous decision
[00:12:10] making. helping them via teleoperation was allowed in some
[00:12:14] cases, but penalized. so
[00:12:16] robots can be somewhat smart and fast. does that mean they could
[00:12:20] do our chores soon? watching videos like these, you might think
[00:12:24] householder. robots are already working in people's homes, but shipping
[00:12:28] robots is one thing, what they can actually do is another, and
[00:12:31] testing environments, robots might succeed and ironing clothes,
[00:12:35] folding laundry or doing the dishes. everyday
[00:12:38] homes are more complicated. our homes aren't
[00:12:41] standardized. it's extremely difficult for
[00:12:44] humanoid sensors to understand everyday household
[00:12:47] chaos. For now, the idea of humanoid robot walking into your
[00:12:51] home and cleaning it from top to bottom without any...
[00:12:55] resistance remains science fiction. People think that
[00:12:58] maybe we can just put llams on robots and suddenly they'll be
[00:13:01] intelligent, but the lm are not
[00:13:05] really intelligent in the sense that they don't understand the
[00:13:08] physical world and they don't understand consequences of
[00:13:11] actual. One of the biggest technical challenges is
[00:13:14] touch or tactile sensing. Robots have
[00:13:17] touch and force sensors, but they still cannot
[00:13:20] match the sensitivity of human touch. wang shing
[00:13:24] shing see. of Chinese company Unitrie recently admitted that
[00:13:28] humanoid robots could still need up to 10 years before
[00:13:32] they have a real breakthrough. Okay, but what about humanoids in
[00:13:35] the workplace? Will they take our manual jobs?
[00:13:39] Companies are already testing humanoid robots in factories,
[00:13:43] warehouses and dangerous work environments. They bring
[00:13:46] one massive advantage from the employer's perspective. They're built to
[00:13:50] work in spaces designed for humans. That could mean
[00:13:53] fewer changes to... factories and equipment, but here's
[00:13:57] the good news for workers: the World Economic Forum expects robot
[00:14:01] technology to create more jobs than it will replace.
[00:14:05] While humanoid robots may be stealing the spotlight, the
[00:14:08] bigger threat to the way we work today is actually software
[00:14:11] AI. Yet, the rapid progress in robotics
[00:14:14] raises another important question: what does it mean when the
[00:14:18] world's robots come from China? China doesn't only have
[00:14:21] the flashiest robots at the moment, it's also the global... leader and
[00:14:25] mass producing them. In 2025, Chinese made
[00:14:29] humanoid robots accounted for around 85%
[00:14:32] globally. There's a new tech race taking shape. The US is moving
[00:14:36] to restrict and ban imports of Chinese autonomous humanoids, largely over
[00:14:40] security concerns. Other countries are watching closely
[00:14:43] too. These robots collect and process large
[00:14:47] amounts of data about the people and places around them.
[00:14:51] That raises privacy and national security concerns. the
[00:14:55] bottom line, we likely won't see a Chinese robot invasion
[00:14:58] in our living rooms anytime soon, but physical AI could still
[00:15:02] transform the workplace, and that's something none of us can afford
[00:15:06] to ignore.
[00:15:09] Our AI companies really destroying antique
[00:15:13] books, AI companies are burning books, burning the books. Big
[00:15:16] AI is literally lighting our books on fire. You've probably seen videos
[00:15:20] like these on social media, but it's not that
[00:15:23] simple. Here's what's... Anthropic, one of the biggest
[00:15:27] AI companies in the world ran secret project, their
[00:15:30] own internal documents state and effort to destructively
[00:15:34] scan all the books in the world, and they don't want any... want to know
[00:15:37] about it, the optics aren't great,
[00:15:41] but are they really destroying unique cultural
[00:15:44] treasures? let's take a closer look at what's actually happening. first
[00:15:47] question, why? AI models like chat chipt or
[00:15:51] cloud learn from huge amounts of text. for
[00:15:54] years, companies just scraped the internet for training
[00:15:58] data. they took everything they could find, but even the
[00:16:01] internet is finite. by now, there's not much fresh,
[00:16:05] high quality text left to feed. these models,
[00:16:10] the internet is now being written by AI itself, and when an
[00:16:14] AI trains on AI generated content,
[00:16:17] quality drops fast. researchers call it model
[00:16:20] collapse. think of it like photocying a photocopy
[00:16:24] of a photocopy. it gets worse every time. That's where
[00:16:28] books matter, specifically older ones written long before
[00:16:31] the AI era. They're curated, reviewed, structured,
[00:16:35] and arguably, the highest quality training data left on the
[00:16:39] planet. In recent months, booksellers, researchers and
[00:16:42] distributors began noticing large scale purchases of physical
[00:16:46] books, like this antiquarian book shop in Berlin. We were
[00:16:50] getting orders worth over €100 from this company,
[00:16:53] almost every day, and some of them would come in at
[00:16:56] 4:30, 5, 5:30, so they weren't
[00:17:00] bundled together, which suggests these had to be
[00:17:02] automated orders. There's one more reason why
[00:17:06] Anthropic is by. physical books recently, they agreed to
[00:17:10] huge $1.5 billion dollar settlement in a
[00:17:13] copyright case. a group of authors had sued the company for training their
[00:17:17] AI models on their texts, and here's the interesting part: the court
[00:17:21] found that training AI on books is not
[00:17:23] illegal, but downloading pirated copies and storing
[00:17:27] them is. so now anthropic is paying for the hard
[00:17:31] copies, and in this case it was ruled that buying books,
[00:17:34] digitizing them and destroying physical copies is... legal
[00:17:37] under US law, it's only changing the format. Are
[00:17:41] they actually destroying rare, unique books that will now be lost
[00:17:45] forever? In short, so far, there's no evidence of it.
[00:17:49] Anthropic mostly looks for out of print, technical, or
[00:17:52] super niche
[00:17:55] the books ordered were from the 70s, 80s,
[00:17:58] 90s. From that time, many books still haven't been digitized
[00:18:02] to this day, meaning they're not freely available online or
[00:18:05] not available on the internet. at all, and it was mainly
[00:18:09] non-fiction and specialist books, so specific
[00:18:12] subjects with a high information density and academic style
[00:18:16] of language, the reality, most used books
[00:18:20] aren't rare, they exist in thousands of identical
[00:18:23] copies. Nevertheless, henwack stopped selling the books when they
[00:18:27] realized what they were being used for because they wanted to sell books for
[00:18:30] humans to read. Other booksellers are happy they get to sell
[00:18:34] stacks of old books that have been sitting in their warehouses for years. but
[00:18:38] critics worry AI companies won't stop at mass produced books
[00:18:41] and will come for the rare ones
[00:18:43] next.
[00:18:46] Mark Zuckerberg recently dropped Meta's version for the future of
[00:18:49] AI, a personal superintelligence designed to help
[00:18:53] all of us reach our goals. He also warns against concentrating
[00:18:56] AI power in the hands of few. has Zuckerberg suddenly
[00:19:00] become a tech humanitarian, don't be fooled. behind his
[00:19:03] massive 6,500 word manifesto is much b
[00:19:07] strategy, a high stakes battle for influence, data
[00:19:11] and money. Here's what's really going on. Let's first take a look at the
[00:19:15] huge promises he's making. Zuckerberg argues that
[00:19:18] individual empowerment must be at the center of AI
[00:19:22] development, and according to him, there's a lot in store for the average
[00:19:26] user. In his vision, we will all have super
[00:19:29] capable personal AI agents that work for us
[00:19:32] 24/7. We'll have incredible tools to get
[00:19:35] creative and powerful tools. to bring our business ideas to life,
[00:19:39] plus we will be able to learn anything we want and can
[00:19:43] all contribute to scientific progress, and the best part, according to
[00:19:46] Zuckerberg's vision, all of that should be free, or at least
[00:19:49] affordable for billions of people. Sounds great,
[00:19:53] right, but some tech journalists and experts view the whole
[00:19:56] thing as a clumsy attempt to make AI more appealing. Fact
[00:20:00] is, global public trust in AI is low.
[00:20:03] According to a recent study, users are generally more afraid of...
[00:20:07] potential dangers, then they are excited by the possibilities it might
[00:20:11] bring. there is another hidden message in the
[00:20:14] manifesto. Zuckerberg wants to position
[00:20:17] Meta as the good guy in the AI race. Look at how
[00:20:20] companies like Open AI and Anthropic market
[00:20:23] themselves. They warn that AI agents could go rogue or
[00:20:26] cybersecurity is in danger. Then they offered to
[00:20:30] sell us tools designed to help manage those risks.
[00:20:33] Zuckerberg basically calls us out as fear mongering, designed to
[00:20:36] trap. corporate and private users into expensive paid
[00:20:40] subscriptions, and that's where meta's strategy comes into focus:
[00:20:43] unlike many of its AI rivals, meta doesn't depend
[00:20:47] primarily on selling AI subscriptions. it's main
[00:20:50] business driver is still advertising. That means Meta can
[00:20:54] afford to offer powerful AI tools for
[00:20:57] free, or a much lower price than competitors. And if
[00:21:01] users don't have to pay a monthly fee, companies that rely on
[00:21:04] subscriptions could feel the pressure. But Zuckerberg doesn't
[00:21:07] stop there. He even wants to make Meta's AI products open
[00:21:11] source or open weight.
[00:21:14] Let's sum up both as open AI for now though. According to
[00:21:18] Zuckerberg, this is the best. way to help protect individual
[00:21:21] freedom. His sales pitch is simple: if I let
[00:21:25] people look inside my software, they'll trust me far more than
[00:21:29] closed labs like open AI and anthropic. To be fair, meta
[00:21:32] already offers a range of open AI tools, the best known
[00:21:36] being the large language model Lama. But there's a
[00:21:39] geopolitical angle here as well: more and more powerful AI
[00:21:42] models are emerging from China, and Zuckerberg makes it clear that he
[00:21:46] sees that as a strategic challenge. In a chapter.
[00:21:50] titled protecting freedom and preventing government tyranny, Zuckerberg
[00:21:53] warns that if western regulators clamp down on Open AI
[00:21:57] models, autocratic regimes will win the global AI
[00:22:00] race. Critics see it differently. They argue Meta is
[00:22:04] trying to protect its business model, giving away AI
[00:22:07] technology widely, building massive adoption,
[00:22:10] weakening rivals that rely on keeping their most powerful models behind a
[00:22:14] paywall, and keeping regulators at arm's
[00:22:17] length. In summary, Zuckerberg'. really wants
[00:22:21] everyone to trust him, users, governments,
[00:22:24] investors, and especially AI
[00:22:27] regulators, but to many his AI utopia feels
[00:22:30] implausible and self-interested. Let us know what
[00:22:33] you think. See you next
[00:22:47] time. Who justified the enormous salaries? Is there
[00:22:50] principle or among politicians does the saying
[00:22:52] hold true, there are no eternal friends or enemies, only
[00:22:56] eternal interests, who builds, if they destroy the state,
[00:22:59] we monitor the work of institutions, because that's where it all
[00:23:02] begins,
[00:23:09] undoubtedly your trust, your support, we open the most important
[00:23:13] topics, from crime, corruption and constitutional issues, our goal is
[00:23:16] a more just society, a society tailored to every citizen of Bosnia and
[00:23:20] Herzegovina. Through the fight for your rights,
[00:23:24] we always walk with you. In your name, we search
[00:23:27] for answers. With you, we seek
[00:23:30] solutions.
[00:23:34] Your concern for a better, safer environment, the preservation of natural
[00:23:38] resources, the improvement of mutual relations, is our
[00:23:41] focus, because we are also part of Bosnian-Herzegovinian
[00:23:44] society.
[00:30:20] What are this lot doing? What's with all the
[00:30:23] dancing, the lader hosen and the
[00:30:25] beer? And where do they originally hail from? The
[00:30:29] mysterious Barbarians, new archaeological find. show that
[00:30:32] even 1 and a half thous years ago they were sturdily built, well
[00:30:36] nourished and tall.
[00:30:40] This and more on DW's science show, welcome to Tomorrow
[00:30:44] today.
[00:30:50] Untold treasures lie hidden here beneath the
[00:30:54] soil. 4,00 artifacts and remains have
[00:30:57] already been recovered.