Slides: The State of AI Code Quality: Hype vs Reality — Itamar Friedman, Qodo
Source Video
The State of AI Code Quality: Hype vs Reality — Itamar Friedman, Qodo
Relationship To World's Fair 2026
These slides are extracted from a public AI Engineer YouTube video connected to World's Fair 2026. Speaker-matched clips are supporting context unless later confirmed as exact session recordings; official livestream recordings are day-level/event-level source material.
Related Scheduled Sessions
- No individual scheduled session mapping has been assigned yet; treat this as an event livestream deck.
Extracted Slides
OCR text:
opob
The State of Al Code Quality
Hype vs.Reality
ItamarFriedman
Qodo CEOand Co-Founder
GoogleDeepMind
OCR text:
ard 3 Cloud Outages in 7
ot 3 Weoks aoe
7 _ ranere a
a Reantsls
1OY » 20%. > SO.
Related?
65% of developers say at least a quarter of i
@ach connit is generatea or shaped by Al
f ni
il
now Al-influenced.
Naan REALITY THE STATE OF AI CODE QUALITY: H
CP EEERMITEE «ITAMAR FRIEDMAN /co-Founders ceo “PAqodo
OCR text:
Tye
L
allowed-tools description
remote show，Read,Glob，Grep,LS,Task Bashgt diff:,Bash（git status:）,Bash(git log:）,Bash(git show:)，Bash(git on the current branch Completea securtyreview of the pend Sing changes
Youarea seniorsecurityengineerconductingafocusedsecurityreviewof thechangeson this branch
potentlThisisnota generalcodereview-focus ONLY on securitymplicationsnewly added bythis PR.Donot coment nexisting security concerms.
CRITICALINSTRUCTIONS:
2.AVOIDSESktticalisesstecsrlowactfinings 4.EXCLUSiONS:Do NOTreport thefollowingissue types: 1.MINIMIZEFALSEPOSTIVES:Onlyflagissues whereyoure>8O%confident ofactual 3.FOCUS ONIMPACT:Prioriize vulnerabiices that couldlead to unauthorized access,data breaches,orsystemcompromi exploitabilty
AIE/LEAD JALITY:HYPEVS.REALITY THESTATEOF AI
GoogleDeepMind PRESENTED BY ITAMARFRIEDMAN/Co-Founder&CEO opob
OCR text:
- cd + e Ter Part 1 State of Al coding adoption
oy bs 0 etn ge oe eee, Oe ee oe te ee ee ee
: ° 1 Sources @Qcdo, GSorar, GFaros we trey ee te ney eee ee eae
code qualty/reviesy related report
; a ee ee
| Samp ¢ size thousands of developers, millions ee ee ee
F of pul requests, and bilhons of lines of code De
Pe
a
.
e ay
= mn ee . 7 a AR NGHiNG
AIE/\.EAD ANTHROP\C aay RY aera teeta AS ne bacdes he we ot nak
Google DeepMind B @JCURSOR $$ Cognition & FACTORY @Graphite ,neod] Ya
OCR text:
tae feLelo)
-”* < aay
, Learn sa System
, Focused on Guaity
2 Wath Aqentic Guaity
: Agentc Code Gen
Code Cer
“i aa P| Investment
a? Eat
e ta
Engineering the future of Al
OCR text:
5 The Quality Crisis ake hole
ad
vap it Pacyone
. How it Beains
; ; 82-92% Developer Adoption
; 1 Al coae gen tools adoption 's grow ng EEE LONE IE Pa SS
3x Productivity Boost in writing code
Al proauctarty dsesn’t guarantee Leb Mg pepe pe pepe POM pee etn ge
: 2 ce ee ee
Foner he
67% Quality Concerns
' 5 a striaa arn ek MnAes aa carer ase ac aT Cerin Of teams report increased difficu ty mamtaming
“x Y assurance ete OP Ieee riety
; oJ fa
Seed Engineering the future of Al
OCR text:
AIE
OCR text:
. ae . The Catch and Crisis in © qodo
a D 2 ~ ist
. Vibo Coding
Al boosts output, but human review becomes the bottleneck
f More code Velocity metres. 1% Change fram low to hugh Ad adopeion
; . rs More PRs ; 5 / : +91.1%
3 Not :¢ss bugs/issues per ine of code ' wes 7 a
Mhatarelatanreriie) S
i on
ry lines of codes are written faster tagaa meee ee eee
a developer car teow ; ; fants
= mi ——
- S
toot
ame STATE OF AI CODE QUALITY: HYPE VS. REA
oe ee ITAMAR FRIEDMAN /co-founderaceo “AAqodo
OCR text:
a Quality Issues That Modern Engineering Stee
» Teams Face
; ; _ 3 Process-Level Challenges
a ile
mt ing P|
STATE OF AI CODE QUALITY: HYPE VS. REA
CT EEEAMTEE ITAMAR FRIEDMAN {co-founders ceo “BAqodo
OCR text:
il
|
3
|
~~
fae,
OCR text:
yy r cwme feXe[o)
| Code review as a
eR W A Me e1RoL ll elo1e
to improve and govern
code-level & process-level quality
wa ay ¥
Engineering the future of Al
OCR text:
Why is Al Code Review Important?
Moreproductive,higherquality
47%
WhenusingAlcodereviewthere are double qualitygain More productive when using Al codereview
Source:QodoState of Al qodo
AlEngineer Engineering the future of Al
OCR text:
Qodo Code Review Stats
Guardrail that detects critical issues, boosts productivity
Sample size: ~1m PRs
17%
Of PRs contained issue with
high severity (9-10)
80%
When AI review tool is enabled,
80% of PRs don't have any
human comment or review
Source: Qodo State of AI Code Quality Report & SonarSource State of Code
AI Engineer
Engineering the future of AI
OCR text:
The Path Forweare:
| 6 ~
(OTE Wate.) Al is a Tool, Start Today,
Competitive Advantage Not a Solution Iterate Tomorrow
atr RE ROOT Gare OAC + Mult-agent System « Your Context Engine
PU Crna a Reet aL aniy » Code & Tools Governance » Al Code Rovew
2 Document & Freped Best Practices « Build Your Quality-specific MCPs. » Al Quality Workflows
« Monitor
|
Ke (ore lo)
aS
PN Tame STATE OF AI CODE QUALITY: HYPE VS. REAL
Google DeepMind ITAMAR FRIEDMAN / Co-Founder & CEO “A qodo
OCR text:
AIE
Slide-Derived Subjects To Review
Subject extraction uses video title, related session titles/descriptions, transcript context, and OCR text when available. OCR is best-effort and should be reviewed against the embedded slide images.