Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
ワールドカフェI /チューターを改良する / World Café I and Improvi...
Search
Kenji Saito
PRO
April 18, 2026
Technology
390
0
Share
ワールドカフェI /チューターを改良する / World Café I and Improving the Tutors
早稲田大学大学院経営管理研究科「プロンプトエンジニアリング ─ 生成AIの応用」2026 春の第1-2回で使用したスライドです。
Kenji Saito
PRO
April 18, 2026
More Decks by Kenji Saito
See All by Kenji Saito
ポスター発表&デモと総括 / Poster Presentations & Demonstrations and Summary
ks91
PRO
0
29
OpenClaw のインストールとデモ / Installing and Demonstrating OpenClaw
ks91
PRO
0
20
AI とサービス・デザイン / AI and Service Design
ks91
PRO
0
170
ルール・ロール・ツールを創る / Creating Rules, Roles and Tools
ks91
PRO
0
160
コーディング・エージェントの活用 / Using Coding Agents
ks91
PRO
0
47
AGI (Artificial General Intelligence) の論点 / Key Issues in AGI (Artificial General Intelligence)
ks91
PRO
0
34
ワールドカフェ再び、そしてゴール・ルール・ロール・ツール / World Café Revisited, and the Goals-Rules-Roles-Tools
ks91
PRO
0
190
AI の技術 / AI technology
ks91
PRO
0
180
対話による知の拡張と傾聴の理論 / Expansion of Knowledge Through Dialogue and the Theory of Active Listening
ks91
PRO
0
36
Other Decks in Technology
See All in Technology
管理アカウント単一運用からAWS Organizationsに移行するの大変で滅
hiramax
0
240
エンジニアは生成AIと どのように向き合うべきか? ことばの意味という観点から
verypluming
3
200
A Harness for Behaviour: how to get AI to generate code that does what we intend, or "TDD in the age of AI"
xpmatteo
0
420
Kaggle未経験社員をメダリストに育てる「AIドラゴン桜」
lycorptech_jp
PRO
0
560
eBPF Can Do It! A 5-Minute Tour of 5 Real-World PHP Issues Solved with eBPF
egmc
0
140
Agentic Design Patterns
glaforge
0
190
なぜハノーバーメッセに行くべきなのか 〜初参加だから語れること〜
tanakaseiya
0
110
CARTA HOLDINGS エンジニア向け 採用ピッチ資料 / CARTA-GUIDE-for-Engineers
carta_engineering
0
47k
Geek Woman の育ち方 〜コミュニティとAIと〜
chicaco
0
410
Python開発環境にハーネス適用を検討する
yuuka51
1
510
はじめてのAI-DLC
yoshidashingo
2
530
freee-mcpを Local→Remote で出してわかった MCP認可実装のリアル
terara
3
630
Featured
See All Featured
A brief & incomplete history of UX Design for the World Wide Web: 1989–2019
jct
2
380
How to build a perfect <img>
jonoalderson
1
5.5k
Collaborative Software Design: How to facilitate domain modelling decisions
baasie
1
220
GraphQLの誤解/rethinking-graphql
sonatard
75
12k
Product Roadmaps are Hard
iamctodd
PRO
55
12k
Max Prin - Stacking Signals: How International SEO Comes Together (And Falls Apart)
techseoconnect
PRO
0
170
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
21
1.5k
Navigating the moral maze — ethical principles for Al-driven product design
skipperchong
2
370
A Modern Web Designer's Workflow
chriscoyier
698
190k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
367
27k
Measuring & Analyzing Core Web Vitals
bluesmoon
9
830
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
47
8.1k
Transcript
Generated by Stable Image Core × Nano Banana 2 —
AI 2026 3-4 I (WBS : ) 2026 3-4 I — 2026-04-20 – p.1/53
https://speakerdeck.com/ks91/collections/prompt-engineering-2026-spring Discord loglm ( ) 2026 3-4 I — 2026-04-20
– p.2/53
( ) 1 4 13 LLM ( ) • 2
4 13 ( ) • 3 4 20 I ( ) • 4 4 20 ( ) • 5 4 27 6 4 27 7 5 11 8 5 11 9 5 18 II 10 5 18 (1) 11 5 25 (2) 12 5 25 (3) 13 6 1 & 14 6 1 & W-IOI 2026 3-4 I — 2026-04-20 – p.3/53
( 20 ) 1 • 2 • 3 (Windows WSL
) • 4 (macOS Lima ) • 5 (macOS ) • 6 • 7 • 8 9 RPG 10 “September 12th” 11 12 13 ∼ 14 AGI (Artificial General Intelligence) 7 (4/27 ) / (2 ) OK 3,4,5 3 / 2026 3-4 I — 2026-04-20 – p.4/53
( ) / LLM ( ) — 1 : 2
: 2026 3-4 I — 2026-04-20 – p.5/53
2 : I AI 2026 3-4 I — 2026-04-20 –
p.6/53
2026 3-4 I — 2026-04-20 – p.7/53
1. (1) AI (2) 2026 4 16 ( ) 23:59
JST ( ) Waseda Moodle 2026 3-4 I — 2026-04-20 – p.8/53
. . . . . . 87 ( ) 72
(4/17( ) ) ( ) A X 2013 20 (1990 ) ( X ) ↑ A AGI X X X A X A ⇒ A critical reasoning ( ) X A 2026 3-4 I — 2026-04-20 – p.9/53
N AI AI AI AI ×AI ⇒ 2026 3-4 I
— 2026-04-20 – p.10/53
H AI A AI X ⇒ X > A X
2026 3-4 I — 2026-04-20 – p.11/53
T chatgpt AI ⇒ AI ( ) AI ( )
2026 3-4 I — 2026-04-20 – p.12/53
W [ ] Open Claw AI ⇒ ( ) 2026
3-4 I — 2026-04-20 – p.13/53
( ) (part 1) 1 2026 3-4 I — 2026-04-20
– p.14/53
LLM “ ” “ ” “ ” AI “ ”
“ ” (a) / (b) (c) “ ” ⇒ 2026 3-4 I — 2026-04-20 – p.15/53
PAT (Claude) LLM LLM LLM A AI X GPT 3
— — AI AI × AI > AI LLM LLM ⇒ ( ) 2026 3-4 I — 2026-04-20 – p.16/53
PAT (GPT) LLM LLM AI AI ⇒ 2026 3-4 I
— 2026-04-20 – p.17/53
( ) (part 2) 2026 3-4 I — 2026-04-20 –
p.18/53
LLM “ ” “ ” → → → AI “
” LLM “ ” ⇒ 2026 3-4 I — 2026-04-20 – p.19/53
PAT (Claude) LLM LLM — LLM LLM — LLM LLM
— ⇒ 2026 3-4 I — 2026-04-20 – p.20/53
PAT (GPT) LLM LLM AI LLM ⇒ 2026 3-4 I
— 2026-04-20 – p.21/53
AI https://www.technologyreview.jp/s/380931/why-opinion-on-ai-is-so-divided/ https://www.technologyreview.com/2026/04/13/1135720/why-opinion-on-ai-is-so-divided/ 2026 3-4 I — 2026-04-20 – p.22/53
AI AI 2026 AI 73% 23% 50 AI AI (
) Claude Code 6 Claude 2 ↑ 2026 3-4 I — 2026-04-20 – p.23/53
(AI) ( ) ( ) AI ( AI ) →
( ) → ( ) AI ( ) GPT-5 Codex OpenAI ( ) ( ) ( ) ( ) >AI ×AI 2026 3-4 I — 2026-04-20 – p.24/53
( ) 6 11 ( ) ( . . .
. . . fix ) 2026 3-4 I — 2026-04-20 – p.25/53
2 : ( ) ID team< >- : team3-interpreter (
) 2026 3-4 I — 2026-04-20 – p.26/53
(agent id ) agent_id team< >- : team3-interpreter . .
. /chat team3 3 2026 3-4 I — 2026-04-20 – p.27/53
(/agent-import ) /agent-import file: ( ) Windows (Notepad) macOS &
( ID ) agent hub & OK ID /import /agent-import file: overwrite: True / overwrite ( ) 2026 3-4 I — 2026-04-20 – p.28/53
2026 3-4 I — 2026-04-20 – p.29/53
I AI 2026 3-4 I — 2026-04-20 – p.30/53
◦◦ vs. ◦◦ 1,000 ( ) 500 16 AI 3
2026 3-4 I — 2026-04-20 – p.31/53
(1) 1. (4-5 ) ( Zoom ) (20∼30 ) (
Discord ) ( ) 2026 3-4 I — 2026-04-20 – p.32/53
(2) 2. 2026 3-4 I — 2026-04-20 – p.33/53
(3) 3. AI 2026 3-4 I — 2026-04-20 – p.34/53
(4) 4. 4 2026 3-4 I — 2026-04-20 – p.35/53
( 1 ) (graceful stop) ∼ 60 ( ) 2026
3-4 I — 2026-04-20 – p.36/53
Zoom Discord ( ) Discord 2026 3-4 I — 2026-04-20
– p.37/53
: 5 : ( ) 11 = ( ) (
) ( ) AI 2026 3-4 I — 2026-04-20 – p.38/53
= /chat [openai_responses] # ACK (Acknowledged; ) ACK ( )
[ROUND_END] ( ) [NEW_ROUND] ( ) 2026 3-4 I — 2026-04-20 – p.39/53
1 (15 ) (0 ) ( ) 1 : /
/ / / / 2026 3-4 I — 2026-04-20 – p.40/53
[ROUND_END] → ( ) 0 n(11?) 1 ( ) 2026
3-4 I — 2026-04-20 – p.41/53
2 (15 ) 5 1 [NEW_ROUND] OK 2 : (
) / / / / / 2026 3-4 I — 2026-04-20 – p.42/53
[ROUND_END] → ( ) 2 0 n(11?) 1 ( )
2026 3-4 I — 2026-04-20 – p.43/53
3 (15 ) 5 2 [NEW_ROUND] OK 3 : (
) / / / / / 2026 3-4 I — 2026-04-20 – p.44/53
[ROUND_END] → ( ) 0 ( ) 2026 3-4 I
— 2026-04-20 – p.45/53
4 (15 ) 5 3 [NEW_ROUND] OK 4 : instructions
AI AI / / / / / 2026 3-4 I — 2026-04-20 – p.46/53
[NEW_ROUND] 4 2026 3-4 I — 2026-04-20 – p.47/53
: , , , 2009 2026 3-4 I — 2026-04-20
– p.48/53
AI GAMER PAT gpt-5.4-nano OK /agent-show-full 2026 3-4 I —
2026-04-20 – p.49/53
2026 3-4 I — 2026-04-20 – p.50/53
2026 3-4 I — 2026-04-20 – p.51/53
2. (1) AI AI AI OK (2) 2024 4 23
( ) 23:59 JST ( ) Waseda Moodle 2026 3-4 I — 2026-04-20 – p.52/53
2026 3-4 I — 2026-04-20 – p.53/53