The End Of Knowledge - Vault 7/366 - xHubAI - 20/08/2025 - 🔴PROGRESO DE LA AGI ： Avances sorprendentes y el camino por delante

graph LR classDef openai fill:#ffd4a3,font-weight:bold,font-size:14px; classDef talent fill:#a3d4ff,font-weight:bold,font-size:14px; classDef europe fill:#a3ffd4,font-weight:bold,font-size:14px; classDef culture fill:#ffa3d4,font-weight:bold,font-size:14px; classDef safety fill:#d4a3ff,font-weight:bold,font-size:14px; classDef future fill:#ffffa3,font-weight:bold,font-size:14px; Main[Vault7-366] Main --> P1[Jakub leads
after Ilya exit. 1] P1 -.-> G1[OpenAI] Main --> P2[IMO gold
but limits remain. 2] P2 -.-> G1 Main --> P3[Models know
when they fail. 3] P3 -.-> G1 Main --> P4[Teacher trained
future OpenAI stars. 4] P4 -.-> G2[Talent] Main --> P5[Human mentors
beat AI tutorials. 5] P5 -.-> G2 Main --> P6[AGI redefined
as discovery engine. 6] P6 -.-> G1 Main --> P7[Scaling compute
time unlocks breakthrough. 7] P7 -.-> G1 Main --> P8[Utility beats
saturated benchmarks. 8] P8 -.-> G1 Main --> P9[Model second
to human Saiho. 9] P9 -.-> G1 Main --> P10[Late-night shock
when reasoning worked. 10] P10 -.-> G1 Main --> P11[GPT-4 insight
sparked Sidor AGI. 11] P11 -.-> G1 Main --> P12[GDP lags
capability curves. 12] P12 -.-> G3[Economy] Main --> P13[EU elite
destroy history. 13] P13 -.-> G4[Europe] Main --> P14[Spain talent
flees lack vision. 14] P14 -.-> G4 Main --> P15[AI aids
OpenAI alignment. 15] P15 -.-> G5[Safety] Main --> P16[GPT-5 hinted
no details. 16] P16 -.-> G1 Main --> P17[Genius must
learn communication. 17] P17 -.-> G2 Main --> P18[Super-chats fund
Spanish AI. 18] P18 -.-> G6[Culture] Main --> P19[Summer spikes
Discord queues. 19] P19 -.-> G6 Main --> P20[Friday debate
Zuckerberg strategy. 20] P20 -.-> G6 Main --> P21[HRM paper
pre-recorded review. 21] P21 -.-> G6 Main --> P22[Selfies Leger
episode 3. 22] P22 -.-> G6 Main --> P23[2025 agents
of art. 23] P23 -.-> G7[Future] Main --> P24[Europe warned
of sodomización. 24] P24 -.-> G4 Main --> P25[China hidden
model threat. 25] P25 -.-> G8[Geopolitics] Main --> P26[UBI linked
to birth collapse. 26] P26 -.-> G3 Main --> P27[Immigration seen
as invasion. 27] P27 -.-> G4 Main --> P28[Host embraced
public speaking. 28] P28 -.-> G6 Main --> P29[Code teaches
problem decomposition. 29] P29 -.-> G2 Main --> P30[GPT-4 trusted
with Gmail. 30] P30 -.-> G1 Main --> P31[Pop culture
sparked AI careers. 31] P31 -.-> G2 Main --> P32[Essay inspired
Polish teens. 32] P32 -.-> G2 Main --> P33[AlphaGo ended
skepticism. 33] P33 -.-> G1 Main --> P34[Politicians ignore
systemic fires. 34] P34 -.-> G4 Main --> P35[Coffee funds
Spanish AI. 35] P35 -.-> G6 Main --> P36[ASI safety
frameworks ahead. 36] P36 -.-> G5 Main --> P37[Summer slowed
for beach. 37] P37 -.-> G6 Main --> P38[Replay on
KissHabby comedy. 38] P38 -.-> G6 Main --> P39[Five years
Spanish AI. 39] P39 -.-> G6 Main --> P40[Engineering channel
teased September. 40] P40 -.-> G6 Main --> P41[XHAVAY TV
backup plan. 41] P41 -.-> G6 Main --> P42[Live loyalty
despite heat. 42] P42 -.-> G6 Main --> P43[3 a.m.
heat commitment. 43] P43 -.-> G6 Main --> P44[Friday delay
insomnia reason. 44] P44 -.-> G6 Main --> P45[No tools
pure reasoning. 45] P45 -.-> G1 Main --> P46[ChatGPT stats
honest metric. 46] P46 -.-> G1 Main --> P47[Public fears
plateau wrongly. 47] P47 -.-> G3 Main --> P48[Automate AI
research safely. 48] P48 -.-> G5 Main --> P49[Engineers need
avatars storytellers. 49] P49 -.-> G2 Main --> P50[Reflect before
AI decides. 50] P50 -.-> G7 Main --> P51[Podcast delays
will fix. 51] P51 -.-> G6 Main --> P52[Frog boiled
society metaphor. 52] P52 -.-> G7 Main --> P53[Resist sodomización
tomorrow debate. 53] P53 -.-> G6 G1[OpenAI] --> P1 G1 --> P2 G1 --> P3 G1 --> P6 G1 --> P7 G1 --> P8 G1 --> P9 G1 --> P10 G1 --> P11 G1 --> P15 G1 --> P16 G1 --> P30 G1 --> P33 G1 --> P45 G1 --> P46 G2[Talent] --> P4 G2 --> P5 G2 --> P17 G2 --> P29 G2 --> P31 G2 --> P32 G2 --> P49 G3[Economy] --> P12 G3 --> P26 G3 --> P47 G4[Europe] --> P13 G4 --> P14 G4 --> P24 G4 --> P27 G4 --> P34 G5[Safety] --> P15 G5 --> P36 G5 --> P48 G6[Culture] --> P18 G6 --> P19 G6 --> P20 G6 --> P21 G6 --> P22 G6 --> P28 G6 --> P35 G6 --> P37 G6 --> P38 G6 --> P39 G6 --> P40 G6 --> P41 G6 --> P42 G6 --> P43 G6 --> P44 G6 --> P51 G6 --> P53 G7[Future] --> P23 G7 --> P50 G7 --> P52 G8[Geopolitics] --> P25 class P1,P2,P3,P6,P7,P8,P9,P10,P11,P15,P16,P30,P33,P45,P46 openai class P4,P5,P17,P29,P31,P32,P49 talent class P13,P14,P24,P27,P34 europe class P18,P19,P20,P21,P22,P28,P35,P37,P38,P39,P40,P41,P42,P43,P44,P51,P53 culture class P15,P36,P48 safety class P12,P26,P47 economy class P23,P50,P52 future class P25 geopolitics

Resume:

The broadcast opens with a tongue-in-cheek nod to the heatwave and the host’s coffee, then frames the session as a follow-up to yesterday’s Greg Brockman interview, this time zeroing in on OpenAI’s chief scientist Jakub Pachocki and research colleague Simon Sidor. After a long, emotionally charged monologue about Spanish and European politics—wildfires, corruption, immigration, birth-rate collapse—the host finally cues the English-language podcast in which the two researchers discuss how they measure AI progress, the meaning of AGI, and the recent “reasoning” breakthrough that lets models talk to themselves before answering. Pachocki argues that true impact will arrive when AI can autonomously generate new science and technology, not merely ace exams, while Sidor recalls how a decade of benchmark saturation has forced OpenAI to look beyond IMO medals toward real-world utility and safety. The host, unimpressed by their on-camera charisma, insists that brilliant coders must learn to communicate or be replaced by better storytellers, and closes by urging the audience to support independent coverage, join the summer Discord, and prepare for Friday’s debate on Meta’s AI strategy.

Inside the recorded interview, the OpenAI duo explain that “chief scientist” means setting the long-term research roadmap, recount how a legendary Polish high-school teacher created a pipeline of programming-competition champions, and stress that while ChatGPT can now tutor students, the emotional scaffolding a human mentor provides remains irreplaceable. They date their personal “AGI moment” to different milestones—Sidor’s was GPT-4’s surprise, Pachocki’s was the first model that could admit “I can’t solve this” on IMO problem six—and agree that the next leap will come from letting models spend vastly more compute-time persisting on hard problems, effectively turning them into automated R&D departments that iterate on code, designs and experiments far faster than any human team.

Wrapping up, they advise today’s teenagers to learn coding not for syntax but for structured thinking, to ignore doom-and-gloom headlines because economic impact always lags capability curves, and to remember that perceived constraints—geography, curriculum, social class—are largely illusory. The host returns to Spanish to vent that OpenAI’s communication strategy is “in SOS,” praises Anthropic and Google, warns that China may already possess a hidden model capable of check-mating U.S. industry, and pleads for a new narrative that empowers citizens rather than lulling them into passive “sodomización.” He promises upcoming episodes on hierarchical reasoning, ASI architectures, and agentic art, asks viewers for coffee-shop donations, and signs off with a call to keep the Discord community “on fire” through the summer lull.

Key Ideas:

1.- Jakub Pachocki became OpenAI chief scientist in May 2024 after Ilya Sutskever’s departure.

2.- The IMO gold medal was reached sooner than expected, yet problem six still exposes reasoning limits.

3.- Models now recognize their own failure, a milestone against hallucination and overconfident output.

4.- High-school teacher Mr. Szubartowski in Poland trained multiple future OpenAI researchers via contests.

5.- Emotional mentorship is deemed irreplaceable despite AI’s ability to generate interactive tutorials.

6.- AGI definition shifts from conversational skill to autonomous technological discovery at scale.

7.- Scaling compute-time persistence on hard problems is framed as the clearest next breakthrough.

8.- Benchmark saturation forces OpenAI to value real-world utility over narrow academic metrics.

9.- The AdCutter 10-hour Japanese optimization contest saw OpenAI’s model place second to human rival Saiho.

10.- Internal shock hit OpenAI when reasoning models first worked, prompting late-night readiness talks.

11.- Simon Sidor’s personal AGI moment arrived when GPT-4 surprised him with unexpected insights.

12.- Economic impact percentages lag capability curves, echoing the early-Web invisible-to-GDP effect.

13.- The host claims Europe’s political elite are “destroying history” through weak AI narratives.

14.- Spanish talent exodus is blamed on lack of national tech vision compared with immigration policies.

15.- OpenAI uses its own models to research alignment and safety, calling human-only oversight insufficient.

16.- GPT-5 development is hinted but no timeline or capability details are disclosed.

17.- The host insists technical geniuses must learn public communication or be replaced by better storytellers.

18.- Viewer donations via YouTube super-chats fund independent Spanish-language AI coverage.

19.- Discord community activity spikes in summer, queuing guests for upcoming X-Talks debates.

20.- Friday’s scheduled debate will focus on Meta’s AI strategy and Zuckerberg’s governance choices.

21.- Hierarchical Reasoning Model paper will be reviewed in a pre-recorded episode to reduce live workload.

22.- The Selfies Leger episodic series continues with episode 3, blending fiction and AI commentary.

23.- Host plans a 2025 “agents of art” program exploring creative autonomy in generative systems.

24.- Europe is warned of cultural “sodomización” if it allows U.S. and China to dominate AI narratives.

25.- China is speculated to possess a hidden model capable of collapsing U.S. AI industry overnight.

26.- Birth-rate collapse and universal basic income are linked to automation replacing human labor value.

27.- Immigration is reframed from welcome policy to perceived cultural invasion amid low national birth rates.

28.- The host admits early personal discomfort with public speaking but embraced it for technological impact.

29.- Programming is promoted not for coding jobs but for structured problem-decomposition skills.

30.- GPT-4’s trust threshold is illustrated by letting it read Gmail without fear of Ewok fan-fiction spam.

31.- The Martian and Iron Man are cited as pop-culture triggers for careers in robotics and AI research.

32.- Paul Graham’s “Hackers and Painters” is recalled as inspiring Polish teens to dream bigger.

33.- AlphaGo Zero’s self-play victory is remembered as the moment skepticism about deep learning died.

34.- The host criticizes Spanish politicians for blaming climate-change arson without addressing systemic issues.

35.- Viewers are encouraged to buy coffee or use PayPal to sustain independent AI dissemination in Spanish.

36.- Upcoming content includes analysis of ASI architecture papers and super-intelligence safety frameworks.

37.- The host vows to slow summer output slightly to allow beach time while maintaining guest quality.

38.- Replay of the GPT-5 reveal is recommended on KissHabby channel for its perfect comedic timing.

39.- The program claims five seasons and five years serving Spanish-speaking AI and “extravist” sciences.

40.- A second engineering-focused channel is teased for September, emphasizing practical development and business.

41.- The host proposes an independent platform “XHAVAY TV Plus” to host content if mainstream channels vanish.

42.- The live audience is praised for loyalty despite summer heat and unpredictable broadcast hours.

43.- The host confesses sleeping only after 3 a.m. due to extreme heat, underscoring commitment to the show.

44.- The Friday X-Talks announcement is delayed because the host “can’t sleep” but promises imminent details.

45.- The IMO model competed without calculators or external tools, relying purely on internal reasoning steps.

46.- ChatGPT usage stats are proposed as a more honest progress metric than obscure academic benchmarks.

47.- The host warns that if six weeks pass without benchmark gains, the public wrongly assumes AI has plateaued.

48.- OpenAI researchers believe automating AI research itself is among the most important tasks to achieve safely.

49.- The host insists that brilliant engineers who can’t communicate should use avatars or ghostwriters for outreach.

50.- The program ends with a call to reflect on what present and future society wants before AI decides for us.

51.- The host acknowledges thousands of podcast listeners and promises to fix recent Evox publishing delays.

52.- The final metaphor compares society to frogs slowly boiled until unable to react to AI-driven control.

53.- Viewers are invited to tomorrow’s debate and reminded that continued resistance against “sodomización” is vital.

Interviews by Plácido Doménech Espí & Guests - Knowledge Vault built byDavid Vivancos 2025