Ukraine War Update NEWS: Pt 1 - Overnight & Other News
Table of Contents 📖
"My goodness me, imagine how lucky you would feel there with that FPV drone landing at your front door."
Hello Team
🎦 00:00-01:22⏩
*Jonathan welcomes viewers to the Ukraine War News update for 25th January 2025, part 1. *He mentions releasing a video for the previous day but has been unwell with a viral illness, along with others on his trip to Ukraine including Benny Pye and Greg Terry. *Zhenya from Kyiv arranged for medication to be sent to them. *Jonathan describes feeling very unwell, even having to crawl to the toilet due to his MS being affected by the illness. *He jokes about the timing of getting ill while in Ukraine. *He promises to keep the update brief after "rambling on".
Return to top⤴️
🪦 DISCLAIMER FOR GENERAL STAFF LOSSES DATA
- These are real people with real lives and real families who love them. Don’t let the numbers sap your humanity.
- These numbers probably aren’t accurate but they’re the best we have and we don’t need them to be accurate to be indicative of patterns of activity.
- All losses are estimates. Losses cannot be counted with accuracy because of the conditions on the ground.
- Both sides would see it to be of their advantage to minimize their own losses maximize the other side’s losses.
- Neither side releases their losses but we have enough transparency from the Ukrainian side to have confidence in they are indicative.
- Personnel losses are hard to count. If a soldier gets injured, heals up, and returns to the front line only to get injured again, is that one loss or two? Also, how to deal with losses from PMC’s or soldiers fighting with RF from occupied territories?
- Equipment losses are hard to count. If an AA complex involves several parts and one part gets disabled, is that a loss, or a fraction of a loss? If a tank gets disabled, repaired, back into the fight, then disabled again, is that one lost tank or two?
- All recorded losses are vulnerable to multiple reporting. We have already seen numerous cases of multiple drones in the air reporting the same loss from different angles as multiple engagements.
- Losses are not always reported on the same day they occurred. It is frequent that drone losses are reported at least 24 hours after other terrestrial equipment losses. Certain losses may not be reported for days or weeks for military intelligence reasons.
Russian Losses: 1,650 Personnel, 37 AFVs
🎦 01:22-02:27⏩
*Ukrainian General Staff figures state Russian losses of 1,650 personnel the previous day, which Jonathan notes is very high. *Losses also include 7 tanks (around the daily average), and a particularly high number of 37 troop-carrying Armoured Fighting Vehicles (AFVs). *Artillery system losses are below the daily average. *Vehicle and fuel tank losses are high at 79. *Jonathan references Andrew Perpetua's loss list which shows significant Russian losses of civilian vehicles, ATVs and trucks. *Conversely, Ukraine is shown to be losing more combat assets than Russia, despite the overall loss ratio being 2.5 to 1 (Russian to Ukrainian losses).
Return to top⤴️
Ukrainian Combat Asset Losses: Includes Stryker and V3M113
🎦 02:27-03:16⏩
*Ukraine's combat asset losses include T-64 and T-80 tanks, some identified as older Soviet models, though potentially newer iterations. *Three infantry fighting vehicles (Soviet models) and four APCs were lost irretrievably. *Losses include V3M113s and a Stryker, the latter falling into a hole and then being hit by a Vandal drone. Jonathan notes the frequency of such drone strikes and the effectiveness of Vandal drones (quadcopters carrying munitions). *Three Kozak 3 MRAPs and an infantry mobility vehicle were also lost. *Jonathan considers these losses to be not "crazy valuable".
Return to top⤴️
Russian Equipment Losses: Minimal Combat Assets
🎦 03:16-03:43⏩
*Despite Ukraine losing more combat assets in the provided lists, Jonathan assesses it's "not the worst in the world". *Russian losses include a couple of older D-30 howitzers (one destroyed, one damaged), one tank, and three infantry fighting vehicles (BMP-3s and a BMP-1). *Jonathan characterises these Russian equipment losses as "nothing too much to write home about".
Return to top⤴️
Russian Non-Combat Vehicle Losses: ATVs, Civilian Vehicles and Motorcycles
🎦 03:43-04:00⏩
*Russian losses include seven ATVs, golf buggies and quads, and "an awful lot" of civilian vehicles, including motorcycles.
Return to top⤴️
Ukrainian Nemesis Drone Strikes Buk M3 Air Defence System
🎦 04:00-04:31⏩
*A Ukrainian Nemesis drone is shown destroying a Russian Buk M3 air defence system. *Jonathan notes seeing possibly two Buk systems recently targeted by Nemesis drones. *He shows thermal imagery of the strike, though questions if it is actually nighttime, suggesting daytime thermal imagery might be clearer. *He reiterates the Nemesis drone's effectiveness against Russian air defence.
Return to top⤴️
Ukrainian FPV Drone Lands at Russian Dugout
🎦 04:31-04:47⏩
*A Ukrainian fibre optic FPV drone flew to a Russian dugout in Kursk but failed to detonate. *Jonathan emphasizes the luck of the Russian soldiers in the dugout, imagining their relief at the near miss.
Return to top⤴️
Ukraine Using Fibre Optic FPV Drones
🎦 04:47-04:55⏩
*Jonathan finds it interesting that Ukraine is using fibre optic drones, suggesting they are catching up in this technology.
Return to top⤴️
Leopard 1A5 Tank Survives Multiple Drone Strikes but Eventually Disabled
🎦 04:55-05:32⏩
*Christian Teals reports a Ukrainian Leopard 1A5 tank, upgraded with additional protection, survived multiple Russian FPV drone attacks in frontline battles. *Benjamin Tamler suggests the added grills, meshes and reactive armour contributed to its survival. *Despite surviving numerous strikes, 10 low-cost drone hits eventually disabled the tank. *Jonathan is surprised it took 10 drones to disable the tank and notes the sheer volume of drones in use.
Return to top⤴️
Drone Warfare Intensity Highlighted by Bradley and Stryker Crews
🎦 05:32-06:03⏩
*Jonathan refers to a report from Task and Purpose in Kursk showing drone damage to a Bradley and Stryker, both still operational. *Crews of these vehicles highlight the incredible number of drones they face, emphasising the intensity of drone warfare.
Return to top⤴️
Church of St Nicholas Burns Down in Russia
🎦 06:03-06:18⏩
*In the Yucca region of Russia, the Church of St Nicholas, used as a humanitarian aid collection point for the "special military operation", burned down. *Jonathan speculates about possible sabotage and uses the hashtag "#Russiaonfire".
Return to top⤴️
Ukraine Repels Combined Missile and Drone Attack
🎦 06:18-06:29⏩
*Ukrainian air defence repelled a combined Russian missile and drone attack. *Out of 61 Shahed drones launched, 46 were shot down and 15 were suppressed by electronic warfare, resulting in a claimed 100% interception rate.
Return to top⤴️
Ukrainian Navy Shoots Down Kh-59 Cruise Missile with Vampire System
🎦 06:29-07:15⏩
*Russia launched two Kh-59 cruise missiles, both of which were shot down. *The Ukrainian Navy released footage of a ship-mounted Vampire SAM system intercepting a Kh-59 missile in the Odesa region. *This is possibly the first footage of an L-3 Harris Vampire intercepting a Russian cruise missile using an APKWS laser-guided rocket. *This confirms Ukrainian Navy vessels are equipped with Vampire SAM systems.
Return to top⤴️
Vampire Systems: Capabilities and Quantity Supplied
🎦 07:15-07:45⏩
*Vampire systems are versatile and can be used against ground targets from helicopters as well as for air defence. *Jonathan describes the system as technically simple, like a "technical load on the back of an SUV". *Ukraine has reportedly received only 14 Vampire systems, which Jonathan finds frustrating as Kofi Bhaboa suggested BAE Systems can produce 25,000 rockets per year for these systems and should have supplied 140 systems instead of 14.
Return to top⤴️
Cruise Missile Interception with Low-Tech Vampire System
🎦 07:45-07:54⏩
*Jonathan finds it "really interesting" that a cruise missile was shot down with a relatively low-tech Vampire system. *He mentions he might have included this in a military aid video.
Return to top⤴️
Damage in Kyiv, Khmelnytskyi and Cherkassy Despite 100% Interception Claim
🎦 07:54-08:21⏩
*Despite the claimed 100% interception rate of the Russian attack, significant damage occurred, including to an apartment block near Kyiv, and in Khmelnytskyi and Cherkassy. *Officials attribute fires at critical infrastructure in Cherkassy to falling debris from intercepted projectiles. *Jonathan highlights that even when drones/missiles are shot down or disabled, falling wreckage can cause damage.
Return to top⤴️
Kh-59 Missile Fired from Su-57 Fighter
🎦 08:21-08:54⏩
*Tim White reports that the Kh-59 cruise missile was fired from a Russian Su-57 fighter jet, which Jonathan considers significant. *The Su-57 is a fifth-generation fighter that Russia has not significantly used in the war, possibly due to manufacturing issues or concerns about their effectiveness if shot down.
Return to top⤴️
Strikes Inside Russia: Belgorod Explosion and Warehouse Fire near Morozovsk
🎦 09:54-09:49⏩
*Strikes inside Russia are mentioned, though limited information is available. *A large explosion occurred in Belgorod, appearing to be daytime but possibly related to nighttime strikes. *A warehouse burned down overnight near Morozovsk, close to a military airfield in the Rostov region, likely due to a strike.
Return to top⤴️
Ukrainian Strikes Target Russian Oil Refineries
🎦 09:49-10:20⏩
*Anton Gerashenko highlights a "Russian oil refinery bingo" image showing numerous oil refineries in Russia. *Those marked with red crosses indicate facilities that have been damaged, though not necessarily completely destroyed, by Ukrainian strikes. *Jonathan suggests Ukraine is systematically targeting Russian oil refineries.
Return to top⤴️
Russian Aviation Incidents: Plane Crashes and Emergency Landing
🎦 10:20-10:52⏩
*Two Russian planes experienced incidents in the past 24 hours. *An AN-24 rolled off the runway in the Khabarovsk region and crashed into bushes, reportedly due to engine failure during takeoff; no casualties were reported. *A Boeing 737-800 Aeroflot flight from Moscow to Istanbul made an emergency landing in Astrakhan due to technical issues. *Jonathan concludes "it's not safe to fly in Russia".
Return to top⤴️
Russian Tanker Oil Spill in Black Sea: Environmental Catastrophe
🎦 10:52-11:09⏩
*A Russian tanker disaster in the Black Sea spilled over 4,000 tonnes of oil, causing a major environmental catastrophe. *Experts estimate recovery could take 20 years. *Moscow is downplaying the crisis, according to the Centre for National Resistance. *Jonathan expresses his dismay at this environmental damage.
Return to top⤴️
Wrap up
🎦 11:09-11:31⏩
*Jonathan concludes the update, hoping to upload footage from their recent activities in Ukraine if he can get internet access. *He hopes to be back to his usual schedule tomorrow, feeling better after taking medication. *He thanks viewers and says goodbye.
Return to top⤴️
🤖❓ AI Debrief (post task) - anything the AI didn't understand
🤖💭 AI Plan (pre task) - the AI's step by step thought process
- TASK 2 (Topic Titles):
- I read through the transcript from start to finish, identifying natural breaks in topics. I looked for changes in subject matter, location, or type of news being discussed (e.g., Russian losses, Ukrainian actions, strikes inside Russia, etc.).
- I decided to make "Hello Team" topic 1 and "Wrap up" the last topic as per instructions.
- For each identified topic segment, I formulated a concise, specific, and quantified title. I focused on what was being discussed, where, and how much/many if numbers were provided. I tried to avoid vague titles as per the "bad examples".
- I numbered the topics sequentially and created `
` tags for each.
- I mentally checked if the topics covered the entire transcript and if they were granular enough.
- TASK 3 (Topic Timestamps):
- I went back through the transcript, topic by topic.
- For each topic, I noted the timestamp at the beginning of the topic (where the discussion of that topic starts) and the timestamp at the end (where it transitions to the next topic).
- I used MM:SS format for timestamps under an hour.
- I created `
` tags, ensuring the `id` matched the corresponding `topictitle` `id`. - After assigning timestamps to all topics, I quickly scanned the timestamps to make sure they covered the entire video duration without significant gaps and that the time ranges seemed logical. This served as a check for missing topics or incorrect timestamp assignments. I noticed the timestamps flowed correctly from start to finish.
- TASK 4 (Topic Summaries):
- I went topic by topic again.
- For each topic, I re-read the corresponding transcript segment carefully.
- I extracted the key information, focusing on facts, figures, locations, and Jonathan's opinions or insights. I aimed to capture the essence of each topic concisely.
- I used bullet points to structure the summaries for readability, as suggested. I used bold text sparingly to highlight key pieces of information like names of equipment or locations.
- I made sure to use British English and Ukrainian spellings (Kyiv, Odesa).
- I corrected minor spelling errors in the transcript within the summaries to improve clarity. For example, "Blumensoz law" to "Sod's law".
- I created `` tags for each summary, matching the `id` to the `topictitle` and `topicts` ids.
- TASK 5 (Quote):
- I reviewed the entire transcript, looking for a quote that was significant, insightful, or representative of Jonathan's style. I chose "My goodness me, imagine how lucky you would feel there with that FPV drone landing at your front door." as it captures Jonathan's relatable and slightly humorous style while highlighting a key point about the war's intensity.
- I ensured the quote was concise and made sense out of context.
- I corrected punctuation and minor errors in the quote.
- I wrapped the quote in ` ` tags.
- TASK 6 (Queries):
- I reflected on the entire process and the transcript itself. I considered if there were any parts I didn't understand or any ambiguities. In this case, the transcript was relatively clear and straightforward, so I concluded that I understood everything and noted this in the `
🤖❓ AI Debrief (post task) - anything the AI didn't understand
` tag. If I had encountered issues (e.g., unclear phrasing, unknown terms), I would have detailed them in the `
🤖❓ AI Debrief (post task) - anything the AI didn't understand
` section, asking for clarification.
- Final XML Output and Verification:
- I assembled all the XML tags in the specified structure, ensuring the order was correct (`
🤖💭 AI Plan (pre task) - the AI's step by step thought process
`, `
`, `<date>`, `<part>`, then topic blocks, ` <blockquote>`, `<div class="kg-card kg-toggle-card" data-kg-toggle-state=close><div class=kg-toggle-heading><h4 class=kg-toggle-heading-text>🤖❓ AI Debrief (post task) - anything the AI didn't understand</h4><button class=kg-toggle-card-icon><svg id=Regular xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path class=cls-1 d="M23.25,7.311,12.53,18.03a.749.749,0,0,1-1.06,0L.75,7.311"/></svg></button></div><div class=kg-toggle-content><p><span style="white-space: pre-wrap;"> `).</li> <li>I double-checked that all `id` attributes were consistent across `<h2 class="topictitle">`, `<topicts>`, and `<div class="topicsummary">` tags for each topic.</li> <li>I critically reviewed the entire XML output to ensure it adhered to <i>all</i> the instructions in the prompt, especially the tag names (particularly verifying I used `<topicts>` and not a mistyped tag) and the overall XML structure. I made sure no tags were mismatched or missing closing tags.</li> <li>I mentally ticked off each constraint in the prompt one last time to be absolutely certain I had met all requirements.</li> </ul> By following this systematic and detailed process, I aimed to produce an accurate and well-structured XML summary that fulfilled all the user's requirements. The step-by-step approach, along with checks at each stage, helped to minimise errors and ensure a high-quality output. <div class="kg-card kg-toggle-card" data-kg-toggle-state=close><div class=kg-toggle-heading><h4 class=kg-toggle-heading-text>🤖💭 AI Plan (pre task) - the AI's step by step thought process</h4><button class=kg-toggle-card-icon><svg id=Regular xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path class=cls-1 d="M23.25,7.311,12.53,18.03a.749.749,0,0,1-1.06,0L.75,7.311"/></svg></button></div><div class=kg-toggle-content><p><span style="white-space: pre-wrap;"> My thought process to summarise the video transcript and output XML: <ol> <li><b>Understand the Goal:</b> The main goal is to summarise ATP Geopolitics video transcripts for "ATP Distilled", focusing on extracting key information and presenting it in a structured XML format. I need to act as a diligent editor, writing in British English.</li> </ol> <ol> <li><b>Break Down the Tasks:</b> I see six distinct tasks, each with specific rules and XML output requirements. I will tackle these in order.</li> </ol> <ol> <li><b>TASK 1 (Title, Date, Part):</b> This is straightforward extraction from the video title.</li> </ol> <ul> <li>Identify the base title by removing the date and part.</li> <li>Extract the date (YYYYMMDD) and convert to DD/MM/YYYY.</li> <li>Extract the part (single letter after the date).</li> <li>Wrap each in the correct XML tags: `<title>`, `<date>`, `<part>`.</li> </ul> <ol> <li><b>TASK 2 (Topic Titles):</b> This requires reading the transcript and identifying topic breaks.</li> </ol> <ul> <li>I'll read through the transcript and look for shifts in subject matter.</li> <li>I need to ensure topic titles are specific, quantified and granular, avoiding vague titles as per the examples.</li> <li>Remember to include "Hello Team" as the first topic and "Wrap up" as the last.</li> <li>Use Ukrainian spellings.</li> <li>Number topics sequentially and use `<h2 class="topictitle" id=X>`.</li> </ul> <ol> <li><b>TASK 3 (Topic Timestamps):</b> This is about linking timestamps to the topics.</li> </ol> <ul> <li>For each topic, I'll find the start and end timestamps in the transcript.</li> <li>Use MM:SS or HH:MM:SS format as needed.</li> <li>Use `<topicts id=X>` tags, ensuring the `id` matches the `topictitle` id.</li> <li>Double-check for gaps in timestamps to ensure all content is covered.</li> </ul> <ol> <li><b>TASK 4 (Topic Summaries):</b> This is the core summarisation task.</li> </ol> <ul> <li>For each topic, read the corresponding transcript section again.</li> <li>Summarise the key points, including context, sources (like Andrew Perpetua), Jonathan's opinions.</li> <li>Use bullet points/markdown for clarity.</li> <li>Maintain British English and Ukrainian spellings.</li> <li>Use `<div class="topicsummary" id=X>` tags.</li> </ul> <ol> <li><b>TASK 5 (Quote):</b> Choose a quote that is meaningful or representative of Jonathan's style.</li> </ol> <ul> <li>Review the transcript for a suitable quote.</li> <li>Ensure it makes sense out of context, is concise and well-punctuated.</li> <li>Use ` <blockquote></blockquote> ` tags.</li> </ul> <ol> <li><b>TASK 6 (Queries):</b> Identify any uncertainties or unclear points.</li> </ol> <ul> <li>After completing all tasks, reflect on the process and the transcript.</li> <li>If anything was unclear, note it down. If everything is clear, state that.</li> <li>Use `<div class="kg-card kg-toggle-card" data-kg-toggle-state=close><div class=kg-toggle-heading><h4 class=kg-toggle-heading-text>🤖❓ AI Debrief (post task) - anything the AI didn't understand</h4><button class=kg-toggle-card-icon><svg id=Regular xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path class=cls-1 d="M23.25,7.311,12.53,18.03a.749.749,0,0,1-1.06,0L.75,7.311"/></svg></button></div><div class=kg-toggle-content><p><span style="white-space: pre-wrap;"> </span></p></div></div> ` tags.</li> </ul> <ol> <li><b>XML Structure:</b> Finally, assemble all the XML components in the correct order as specified in the "XML STRUCTURE" example. Crucially, double check that every opening tag has a corresponding closing tag, and especially that I use `<topicts>` correctly and not misstype it as `<div class="topicsummary">` for the timestamp tags.</li> </ol> By following these steps methodically, and paying attention to the rules and examples for each task, I can generate the required XML summary effectively. I will focus on accuracy, detail in topic titles and summaries, and correct XML formatting. I will also be mindful of British English and Ukrainian spellings throughout. The iterative nature of going back to the transcript for each task (titles, timestamps, summaries) will help ensure accuracy and completeness. I will perform a final review before submitting the response to ensure all instructions are followed and the XML is valid. </span></p></div></div></body></div> </div>
