Videos AGkzpxMdPn8

Most Enterprise Agentic Projects Are Doomed, Here's Why — Jess Grogan-Avignon & Jack Wang, Accenture

index_state ready data_status ok

AI Engineer· published 2026-05-28· 0:20:34· en-US· indexed 2026-08-10 19:51

Open on YouTube

57 shot(s).

keyframes kept every frame deduplicated

What was stored

cues: 222; whisperx 222
chunks: 36; from 222 cues
keyframes: 42; kept of 57 captured
frames with text: 42; 734 lines read
chapters: 0; from the source metadata
keyframe bytes: 7.3 MB; word timings on 222 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage	state	model	started	took
`fetch`	done	—	2026-08-10 14:39	2m 55s
`stt`	done	—	2026-08-10 14:42	21s
`chunk`	done	—	2026-08-10 14:42	0s
`text_embed`	done	—	2026-08-10 19:51	0s
`keyframe`	done	—	2026-08-10 14:42	1m 48s
`ocr`	done	—	2026-08-10 14:44	15s
`frame_embed`	done	—	2026-08-10 19:51	8s

Frames, and what the machine read

0:03 #0 done2 line(s)

shot 0·sharpness 658.4
1. AlEngineer0.98
2. EUROPE1.00
0:08 #1 done2 line(s)

shot 1·sharpness 827.9
1. PRESENTINGSPONSOR1.00
2. Google DeepMind1.00
0:13 #2 done3 line(s)

shot 2·sharpness 907.7
1. PLATINUM SPONSORS0.98
2. # Braintrust0.96
3. WorkOS OpenAI0.95
0:21 #3 done8 line(s)

shot 3·sharpness 569.5
1. AlEngineer0.98
2. EUROPE1.00
3. Most enterprise agentic1.00
4. projects are doomed –0.97
5. here's why0.97
6. accenture1.00
7. AlEngineer0.97
8. EUROPE1.00
0:30 #4 done7 line(s)

shot 4·sharpness 5076.7
1. Our world1.00
2. The world of large enterprises1.00
3. AIE1.00
4. accenture0.99
5. AlEngineer1.00
6. Google DeepMind0.97
7. 1260.74
0:54 #5 done7 line(s)

shot 5·sharpness 5321.8
1. Our world0.99
2. The world of large enterprises1.00
3. Control1.00
4. AIE1.00
5. accenture0.99
6. AlEngineer1.00
7. Engineering the future of Al1.00
1:08 #6 skipped

shot 6·duplicate of #5
1:27 #7 done12 line(s)

shot 7·sharpness 2898.9
1. The reality1.00
2. 'Al Achievers'1.00
3. AIE1.00
4. ★1.00
5. ★1.00
6. 66666666660.52
7. accenture1.00
8. Copyright @ Accenture 2026. All rights reserved.0.97
9. AlEngineer1.00
10. AlEngineer0.97
11. 20261.00
12. EUROPE1.00
1:44 #8 done12 line(s)

shot 8·sharpness 3605.2
1. The reality1.00
2. 'Al Achievers'1.00
3. AIE1.00
4. 88%1.00
5. ★1.00
6. Of companies remain stuck.1.00
7. accenture1.00
8. Copyright @ Accenture 2026. All rights reserved.0.96
9. AlEngineer1.00
10. AlEngineer0.97
11. 20261.00
12. EUROPE1.00
1:56 #9 done9 line(s)

shot 9·sharpness 2544.0
1. redefine1.00
2. AIE0.99
3. the1.00
4. game1.00
5. accenture0.99
6. Copyright @ Accenture 2026. All rights reserved.0.97
7. AlEngineer1.00
8. AlEngineer0.98
9. EUROPE1.00
2:27 #10 done8 line(s)

shot 10·sharpness 584.2
1. 5enterpr0.98
2. nsions to predict t0.97
3. fate of your ext agentic project1.00
4. AlEngineer0.98
5. EUROPE1.00
6. accenture1.00
7. AlEngineer1.00
8. EUROPE1.00
2:38 #11 done10 line(s)

shot 11·sharpness 1561.8
1. 5 enterprise tensions to predict the0.99
2. fate of your next agentic project1.00
3. AIE1.00
4. ★1.00
5. ★1.00
6. accenture0.97
7. Copyright @ Accenture 2026. All rights reserved.0.97
8. AlEngineer1.00
9. Engineering the future of Al1.00
10. 20260.99
3:00 #12 done26 line(s)

shot 12·sharpness 2757.1
1. 5 enterprise tensions to predict the1.00
2. fate of your next agentic project1.00
3. AIE1.00
4. ★1.00
5. Speed1.00
6. Value1.00
7. Delivery1.00
8. Trust1.00
9. Moat1.00
10. Human OS for0.98
11. Finance wired0.98
12. Treating1.00
13. Success1.00
14. Static without1.00
15. human pace0.95
16. for certainty1.00
17. agents like1.00
18. defined by0.99
19. compounding1.00
20. features1.00
21. completion1.00
22. accenture1.00
23. Copyright © Accenture 2026. Al rights reserved.0.98
24. AlEngineer1.00
25. Engineering the future of Al1.00
26. 20261.00
3:15 #13 done28 line(s)

shot 13·sharpness 4663.1
1. 010.98
2. Enterprise speed, really?1.00
3. The human operating system0.99
4. python0.99
5. # enterprise_hell.py 1 O(=) complexity0.96
6. DONE = False1.00
7. ★0.93
8. AIE1.00
9. def steerco():1.00
10. global DONE0.96
11. print("Great slides! Redo slides 3, 7, and 14.")0.97
12. print("'Let's align offline and reconvene next week.")0.97
13. DONE = False # always0.96
14. def process():1.00
15. print("Reworked slides. Chased decisions. Aligned 47 stakeholders.")0.98
16. DONE = False $ see: steerco()0.96
17. week = 00.96
18. while not DONE:1.00
19. week += 10.94
20. print(f"\n Week (week) | Budget burned: £(week " 47_500:,)")0.95
21. process()1.00
22. steerco()1.00
23. print(Done.") unreacnable0.96
24. accenture1.00
25. Copyright @ Accenture 2026. Al rights reserved.0.97
26. AlEngineer1.00
27. Engineering the future of Al0.98
28. 20261.00
3:55 #14 done31 line(s)

shot 14·sharpness 4674.5
1. 010.98
2. Enterprise speed, really?1.00
3. The human operating system0.99
4. python0.96
5. # enterprise_hell.py | O(=) complexity0.96
6. DONE = False0.99
7. ★1.00
8. AIE1.00
9. def steerco():0.99
10. global DONE0.95
11. print("Great slides! Redo slides 3, 7, and 14.")0.97
12. ★1.00
13. ★1.00
14. print(""Let's align offline and reconvene next week.")0.98
15. DONE = False # always0.97
16. def process():0.97
17. print("Reworked slides. Chased decisions. Aligned 47 stakeholders.")0.97
18. DONE = False # see: steerco()0.97
19. week = 00.94
20. while not DONE:0.99
21. print(f"\n Week (week) | Budget burned: £(week " 47_500:,)")0.95
22. week += 10.97
23. process()1.00
24. steerco()1.00
25. print(-Done.") unreacnable0.94
26. accenture1.00
27. Copyright @ Accenture 2026. Al rights reserved.0.95
28. AlEngineer0.99
29. AlEngineer0.97
30. 0261.00
31. EUROPE1.00
4:26 #15 skipped

shot 15·duplicate of #13
4:47 #16 skipped

shot 16·duplicate of #13
5:16 #17 skipped

shot 17·duplicate of #13
5:47 #18 done13 line(s)

shot 18·sharpness 1576.9
1. 010.99
2. Pathway to machine speed1.00
3. The supply of code is growing. Approval infrastructure hasn't kept up1.00
4. AIE1.00
5. ★1.00
6. ★1.00
7. ★0.99
8. accenture0.98
9. Copyright @ Accenture 2026. Al rights reserved.0.96
10. AlEngineer0.99
11. AlEngineer0.97
12. 20260.98
13. EUROPE1.00
6:12 #19 skipped

shot 19·duplicate of #18
6:37 #20 done13 line(s)

shot 20·sharpness 655.6
1. Pathway1.00
2. chine speed0.98
3. The supply of code is growing. Approval infrastructure h1.00
4. AlEngineer0.99
5. EUROPE1.00
6. Al has exposed how expensive today's processes1.00
7. are-in time and money1.00
8. The real tech debt goes beyond legacy code0.99
9. Every human process now needs to1.00
10. become executable, adaptive code0.99
11. accenture1.00
12. AlEngineer0.99
13. EUROPE1.00
6:50 #21 done24 line(s)

shot 21·sharpness 5836.9
1. 010.97
2. GitHub Commits1.00
3. Pathway to machine speed1.00
4. 2026 (forecast)1.00
5. 14bn1.00
6. The supply of code is growing. Approval infrastructure hasn't kept up1.00
7. *★*0.53
8. AIE1.00
9. Al has exposed how expensive today's processes0.99
10. are - in time and money0.96
11. ★1.00
12. ★1.00
13. ★1.00
14. The real tech debt goes beyond legacy code1.00
15. Every human process now needs to1.00
16. become executable, adaptive code1.00
17. This isn't new, or about removing control.1.00
18. But it is a major mindset shift0.98
19. 20251.00
20. 1bn1.00
21. accenture1.00
22. AlEngineer0.99
23. Engineering the future of Al1.00
24. 20261.00
7:12 #22 done13 line(s)

shot 22·sharpness 1743.5
1. 020.99
2. A business case for?0.96
3. Typical business cases assume 3 things are knowable0.99
4. upfront. Agentic Al breaks these assumptions0.99
5. *★★0.72
6. AIE1.00
7. ★1.00
8. ★0.99
9. accenture0.97
10. Copyright @ Accenture 2026. Al rights reserved.0.97
11. AlEngineer1.00
12. Engineering the future of Al0.99
13. 20261.00
7:41 #23 skipped

shot 23·duplicate of #22

Transcript

222 cues· 2,940 words· 17,107 chars

chunk 0 · 0:15–0:59 · 99 words · 616 chars
0:15 All right.
0:16 Hi, everyone.
0:17 I'm Jess.
0:18 This is Jack.
0:19 We're very excited to be here today.
0:22 So let us tell you about our world.
0:26 So we work in the world of enormous enterprises, so telecoms, utilities, serving entire nations, government, healthcare that you just heard a little bit about, consumer products in your home right now.
0:39 And when you operate at that scale, actions have consequences.
0:44 So a bad deployment, for example, can take down critical national infrastructure.
chunk 1 · 0:48–1:31 · 86 words · 558 chars
0:48 And so over time, these organizations have built structures for this reality.
0:53 Control, process, repeatability, governance, layers and layers of it.
1:01 And this has worked really well, right?
1:02 Like for years, these companies have seen massive successes and growth, but always at human pace.
1:10 Human pace, that is what's shifting.
1:13 We're entering a world at machine speed and transforming everything we know.
chunk 2 · 1:19–1:59 · 84 words · 494 chars
1:19 Our work, our clients, and ultimately, the societies these enterprises underpin.
1:26 Our research showed that 12% of companies reached what we called AI achiever.
1:32 It means that most of the companies are still stuck piloting and spending millions and perhaps not getting too much in their return.
1:40 80%.
1:44 The tragedy is not just wasted spend.
1:46 It's about falling behind in a world that accelerating beyond what they can compute.
chunk 3 · 1:55–2:38 · 104 words · 594 chars
1:55 Many of you here may ship here on Fridays and then roll back on Saturdays.
2:00 A decision you take an afternoon could easily take an enterprise six months or more.
2:05 Octopus, Klarna, Shane, they think that's insane.
2:09 And then they go on and redefine the games themselves.
2:13 Others studied the games, crafted the playbooks, and ran the workshop.
2:18 But they went home.
2:20 We stayed.
2:22 We shipped through the reality.
2:24 And that is our moat.
chunk 4 · 2:27–3:11 · 119 words · 700 chars
2:27 When you stay, you learn things that the slide decks don't warn you about.
2:32 So for example, it's not just about data availability or API availability that impacts AI success.
2:39 It's the entire enterprise scaffold itself, the very thing that has made these companies so successful, which is increasingly becoming the drag, the thing that is holding them back from capturing AI value at scale.
2:53 So we're going to share with you today five enterprise tensions, learning from our experience deploying AI at large enterprises over the last couple of years.
chunk 5 · 3:03–3:43 · 102 words · 589 chars
3:03 And if you understand these five, we think you can probably predict the success of your next AI project before you get started.
3:09 Cool, let's get into it.
3:14 18 months ago, you probably still need to explain why AI mattered, why speed mattered, but that battle's gone.
3:22 The C-levels are convinced.
3:24 CEOs are terrified of being left behind now.
3:27 But yet, the enterprise speed has not really shifted.
3:31 It's not because AI cannot write good code.
chunk 6 · 3:34–4:17 · 96 words · 614 chars
3:34 It is not because our engineers can't solve the context problem.
3:38 I think it's something a lot more deeper.
3:40 It is the actual enterprise scaffolding itself.
3:44 A human operating system that's designed for human and running at a human speed.
3:50 The automation behind every delivery that I just mentioned about, thinking about data access, security reviews, deployment process, most of the enterprises never needed to invest like a tech company.
chunk 7 · 4:05–4:46 · 99 words · 624 chars
4:05 Corporate process balanced with minimal engineering investment with the complimentary of STACO meetings.
4:12 And that is how enterprise runs today.
4:15 Fit for enterprise, fit for human.
4:19 We had the pleasure of delivering agentic solution in a large corp and integrating their centralized AI gateway.

Most Enterprise Agentic Projects Are Doomed, Here's Why — Jess Grogan-Avignon & Jack Wang, Accenture

Scene timeline

What was stored

Provenance

Frames, and what the machine read

Transcript