검열 불가 특성을 이용해서 악의적 프롬프트를 블록체인에 심어두는 사례입니다 요즘 오픈 프로젝트들을 통한 공격들이 늘어나고 있는데, 이런식으로 여러 채널을 이용하면 그 공격을 막기가 더 까다롭겠습니다
👀1

Channel
@feelgoodforum
On this record: Growth · Engagement · What this channel posts · Reactions · Posts · Citations · Cite this entry
29subscribers
+0 since we began measuring on 8 August 2026
Risers and fallers across the register · movement among entries of Under 1,000.
| Telegram ID | -1003230479914 |
|---|---|
| Type | Channel |
| Username | @feelgoodforum |
| Description | 느좋포럼 |
| Created | Between 1 October 2025 and 31 January 2026— estimated from Telegram’s id allocation, not measured. How this range is calculated. |
| First recorded | 11 August 2026 |
| Last confirmed live | 11 August 2026 |
| Measurements held | 2 |
| On Telegram | t.me/feelgoodforum |
| Measured (UTC) | Subscribers | Change |
|---|---|---|
| 11 Aug 2026, 05:16 | 29 | no change |
| 8 Aug 2026, 06:30 | 29 | first reading |
20 posts held, back to 19 July 2026 — the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 1 pageof Telegram’s post history, 20 posts per page.
ERR is average views per post over the last 30 days divided by subscribers, the definition TGStat uses, so this figure is comparable with the one you will see elsewhere. It falls structurally as a channel grows: a high ERR on a small channel and a low one on a large channel describe reach mathematics, not quality. We publish the figure and the sample it came from and pass no verdict on it.
ER is defined industry-wide as (forwards + reactions + comments) ÷ views— note the denominator is views, not subscribers. Telegram’s public web preview carries views and reactions but not forward or comment counts, so the reaction rate above is the reactions term only and is therefore a floor: the true ER for this channel is higher by an amount we have not measured and will not estimate. It is computed over the 5 of 19 measured posts that carry a reaction reading, and over those same posts' views.
| Window | Rolling 30 days · latest post in window 8 August 2026 |
|---|---|
| Posts held | 20 (19 July 2026 – 8 August 2026) |
| Views total | 308 |
| Reactions total | 6 |
| Forwards / comments | not exposed by the public surface — not measured, not estimated |
| Readings taken | 11 Aug 2026, 05:16 UTC |
Views are a single reading per post, taken at the time above. A post published in the last day or two is still accumulating views, which pulls the 30-day average down slightly. That is a property of the standard definition rather than a fault in it, so we keep the definition rather than “correcting” the number into something nobody can reproduce.
Precision. Telegram publishes view counts on its public widget in short form — 8.12K, 3.7M — so any reading at or above 1,000 reaches us rounded to three significant figures, and only counts below 1,000 are exact. Averages and rates derived from them are shown to the same precision rather than to the unit: a figure like 3,701,250 would assert digits nobody measured.
Reaction counts are published per emoji and rounded the same way, so a total below 1,000 is exact and a larger one is a sum that may carry a rounded component from each emoji above 1,000. Because it is a sum, it does not look rounded — read a large reaction total as three significant figures per contributing emoji rather than as the figure it prints.
Lifetime counters from Telegram’s own channel header, read 11 August 2026 — not the date at the top of this page, which is when the subscriber count was last read. Below Telegram’s rounding threshold, so these counts are exact.
6 reactions across 4 posts, in 3 distinct kinds. The most used accounts for 50.0% of them.
| Reaction | Count | Share | Share, drawn |
|---|---|---|---|
| 👀 | 3 | 50.0% | |
| ❤ | 2 | 33.3% | |
| 👍 | 1 | 16.7% |
No sentiment is inferred, and none should be read in. This table is ordered by count and by nothing else. Emoji do not carry stable meaning across languages or communities — 🙏 is thanks in one channel and mourning in another — so we publish which ones were pressed and how often, and pass no judgement on what an audience meant by them.
Precision. Telegram publishes reaction counts per emoji and short-forms each one — 4.34K, 1.2M — so any single kind at or above 1,000 reaches us at three significant figures, and only counts below 1,000 are exact. The shares above are ratios of those figures and carry the same error. This is also why the total here can differ slightly from a reaction total printed elsewhere on the page: both are sums of the same rounded parts, taken over samples with different edges.
Coverage. Reactions were read on 5 of the 20 sampled posts in this sample. Summed by Telegram’s own count on each post — not by adding up the per-emoji breakdown above — those same posts carry 6reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.
Measured over the 20 most recent posts we hold, published 19 July 2026 to 8 August 2026, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.
검열 불가 특성을 이용해서 악의적 프롬프트를 블록체인에 심어두는 사례입니다 요즘 오픈 프로젝트들을 통한 공격들이 늘어나고 있는데, 이런식으로 여러 채널을 이용하면 그 공격을 막기가 더 까다롭겠습니다
👀1
해커들이 악성코드 명령서를 BNB Chain에 올려두기 시작함 보통 악성코드는 해커가 운영하는 서버에서 다음에 뭘 실행할지 지시를 받아옴. 보안업체가 해당 서버나 도메인을 찾아 차단하면 공격 흐름도 끊을 수 있었음. 그런데 마이크로소프트가 발견한 공격은 이 지시문을 BNB Chain 스마트컨트랙트에 저장해두고, 감염된 웹사이트가 블록체인에서 읽어가는 방식. 특정 회사가 관리하는 서버와 달리 블록체인에 올라간 내용은 외부에서 지우기 어렵기 때문에 일반적인 서버 차단만으로 멈추기 힘들어짐. BNB Chain이 해킹된 건 아니고, 누구나 사용할 수 있는 블록체인의 특성을 해커가 악용한 사례. 공격 자체는 가짜 CAPTCHA로 Win+R → Ctrl+V → Enter를 누르게 만드는 방식임. 웹사이트가 인증이나 오류 해결을 이유로 키보드 명령…
여기도 계속된 메타나 내러티브가 필요한 상태인데... "thinking", "agent", "harness", "loop"에 이어 다음으로는 "autonomous"를 바라보고 있는 것 같네요
오픈AI, 2~3개월 내 AI 활용 방식은 또 한 번 진화 최근 내가 보고 있는 결과들을 보면, Codex는 AI를 활용하기 위한 훌륭한 실행 환경(harness)이라는 점이 꽤 분명해지고 있습니다. 하지만 2~3개월만 지나면 지금의 Codex조차 원시적으로 보일 것입니다. 우리는 최첨단 AI를 사용하는 방식이 또 한 번 크게 진화하는 변곡점을 앞두고 있습니다. 차세대 모델은 더 이상 노트북 하나만으로는 충분하지 않습니다. → 차세대 모델 Astra와 같은 장기 실행 AI를 중심으로, AI 활용 방식이 일회성 질의응답에서 자율 작업 에이전트 중심으로 진화할 것임을 시사 https://t.me/Samsung_Global_AI_SW
https://arxiv.org/pdf/2606.28344 PIXELRAG 라는 아이디어: • 웹페이지를 억지로 텍스트로 변환해서 RAG에 넣지 말고 • 사람이 보는 화면 그대로 스크린숏으로 저장하고 검색하는 RAG 시스템 • 문서 검색도 이미지 임베딩으로 하고, 검색된 근거도 OCR이나 텍스트 변환 없이 VLM이 직접 읽게 만든 시스템 자세하게는: • 웹페이지를 스크린숏 타일로 변환 • 페이지 폭은 875픽셀로 고정하고, 세로로 긴 페이지를 1024픽셀 높이의 타일로 잘라서 저장 • 타일당 하나의 2048차원 벡터만 저장 RAG 동작은: • 각 스크린숏 타일을 Qwen3-VL-Embedding-2B로 임베딩 • 사용자 질문은 텍스트로 입력됨 -> 텍스트 질문과 이미지 타일이 동일한 임베딩 공간에 들어가기 때문에 둘 사이의 유사도를 계산…
중요하고 안전/안정을 생각하면 이더리움에 출시하는게 자연스럽죠 그런 역할을 하는 퍼블릭 네트워크라고 생각하면 이더리움이 참 가치가 좋은데
Open USD는 이더리움에 먼저 출시 출처
과거 신경망 연구들이 "더 큰 모델 더 많은 데이터"로 급격히 전환되면서 일반 연구자들이 연구하기 어려워졌던 때가 생각나네요... 2019년 2020년쯤 CV쪽이 특히 그랬던 것 같은데 LLM도 그 등장에서부터 진입장벽이 좀 있었는데, 갈 수록 더 심해지고 있는 것 같습니다. 이제 sLM이라는 이름의 신경망들도 연구하기 어려워진
LLM 논문이 왜 재미없을까 안녕하세요. 늘 한주에 llm논문을 서너개씩은 올리던 방장입니다. 요새는 한주에 하나, 많으면 두개 겨우 올립니다. 제가 지금까지 공유드린 논문이 50개 가까이가 넘습니다. 논문을 소개드릴때 다음의 기준을 저는 잡고 논문을 분석합니다 -독특하거나 신선한 시도를 하거나 -기존에 다루지 않은 영역/방법론을 소개하거나 -실제 현장,업무에 연결된 실용적인 방법론을 소개/제안/연구 하거나 할때, 그것들을 바로바로 공유드리고자 합니다. 다만 요새 LLM 연구에서, 특히 에이전트와 프롬프팅은 굉장히 지루한(그치만 나쁜건 아닙니다) 상태로 접어들었습니다. [1] 에이전트를 활용해 "이만큼 좋아졌습니다"는 이제 너무 흔합니다. 물론, 모든 연구가 '흔하다' 로 평가절하 당할수는 없습니다. 아직도 몇몇 연구들은 기존에 다…
❤2
요약해보자면: • 핵심은 x402든 뭐든 시스템이 E2E로 전부 자율적으로 동작하고 탈중앙화되어있다는 게 아닌거죠. • 특히 오프체인에서 이뤄지는 액션들에 대한 결과를 온체인에서는 신뢰하기 어렵다보니 (늘 있는 문제입니다만) 정산이라는 행위 자체가 판매자 구매자 모두에게 만족스러운 결과로 정산되는거 자체가 어렵습니다. • 아이러니하게도 이를 해결하는 제일 쉬운 방법은 신뢰할 수 있는 주체가 끼어들어오는것...
3/ 결국 논문으로 돌아가면, 이 공격 벡터들이 나오는 근본 원인은 세 가지 설계 결정으로 나온다고 생각하는데: 첫째, verify가 nonce 소비나 자금 예약 없는 stateless 검사, 사실상 prediction에 그친다는 점 - 카드 결제의 authorization이 자금을 실제로 묶는 상태 변경인 반면, x402의 verify는 "settle 시점에도 유효할 것"이라는 예측일 뿐 보증이 아니므로 취약점 생길 가능성 농후 둘째, 소비자와 머천트를 동시에 보호하려다 보니 verify - execute - settle이라는 분리 구조가 나옴. - 온체인 정산은 되돌릴 수 없으므로 business logic이 성공한 뒤에만 정산해야 하는데, 이 순서가 verify와 settle 사이에 어느 당사자도 책임지지 않는 상태 구간을 만든다. …
2/ https://arxiv.org/html/2607.19545 새로운 논문이 나왔군요 한번 뜯어보죠
Showing the 12 most recent of 20 posts we hold for @feelgoodforum. View and reaction counts are the latest single reading for each post, not a live figure, and a recent post is still accumulating both. A view count marked ≈ was rounded by Telegram before we ever saw it — t.me prints views in full below 1,000 and to three significant figures above, so ≈1,200,000 means somewhere between 1,150,000 and 1,249,999. Unmarked counts are exact. Text is reproduced from the public post preview and truncated for length.
Citation-graph rank — 509,972 of 1,478,351entries in the measured graph. A weighted position computed from the forward and mention edges below — republished posts weigh more than named mentions — and recomputed periodically, over the whole graph. Published only as this ordinal position, never as a score: a position is a fact, and a score printed beside one channel’s name would read as a verdict this register does not make. The two counts beneath stay separate for the same reason mentions are never summed with forwards anywhere else on this page — a named-by count costs nothing to manufacture. The top 100 by this measure, or how it is computed.
Republished by
Channels on the register that have forwarded this channel's posts into their own feed.
Republishes
Channels on the register whose posts this channel has forwarded.
Built only from forwarded posts we have actually read, on both sides. Coverage is early and deliberately incomplete: a missing link means we have not read the post that would prove it, never that the relationship does not exist. Counts are distinct forwarded posts observed, so they only ever go up as we read more.
Named by 1 registered channel — every channel on the register whose own posts have named this one, by its current username or any other username it currently holds, merged from two separately captured readings of the same fact so a namer caught by only one of them is not missed and a namer both caught is not counted twice. A username this channel has since dropped is not matched — that handle may belong to someone else now, and crediting today’s namer to yesterday’s owner would misattribute it.
Named by
Channels on the register whose posts name this channel's handle.
Names
Channels on the register whose handles appear in this channel's posts.
A mention is a weaker signal than a forward and is counted separately for that reason — naming a channel is not republishing it, and a handle in a post body is easy to place deliberately. The post counts beside each row below are distinct posts in which the handle appeared, from posts we have read on both sides — the “Named by N registered channels” figure above is a different count, of distinct NAMING CHANNELS rather than posts, and is not the sum of the rows under it.
A live page changes as we take new readings, so a citation should name the measurement it is based on, not just the URL. The line below cites the subscriber count as measured 11 August 2026 — this entry's latest reading, not the date you are reading this.
“느낌좋은금융마스터포럼” (@feelgoodforum), 29 subscribers as measured 11 August 2026. Telegram Register, tgregister.com/channel/feelgoodforum.
Full measurement history, CC BY 4.0. Every reading this register holds for this entry, not just the latest one, as a dated, downloadable record: CSV · JSON. Free to use with attribution to tgregister.com. Each file carries its own generation timestamp, which is the figure to cite for exactly when the data was retrieved.