Innodata Inc. · NASDAQ:INOD
Innodata — Q3 2023 Earnings Presentation
14 slides from Innodata's earnings presentation, Nov 1, 2023
Slide 1: Q3-'23: Poised for Transformational Growth Driven by AI Wins
Q3-'23 Poised for Transformational Growth Driven by AI Wins with Five of the Largest Global Tech Companies and Leading Enterprises NASDAQ: INOD v1_11072023
Slide 2: Disclaimers
DISCLAIMERS Forward-looking Statements and Exchange Commission. This presentation contains certain forward-looking statements within the meaning of Section Our actual results could differ materially from the results referred to in forward-looking 21E of the Securities Exchange Act of 1934, as amende
Slide 3: We're Excited for Today's Discussion
01. INVESTOR HIGHLIGHTS 05. DIFFERENTIATORS We’re Excited for 02. GENERATIVE AI OPPORTUNITY 06. COMMERCIAL STRATEGY Today’s Discussion 03. HOW WE HELP 07. FINANCIAL OVERVIEW 04. CASE STUDIES 08. MANAGEMENT AND BOARDS 3
Slide 4: About Innodata
01. INVESTOR HIGHLIGHTS About Innodata. We’re a global technology and data engineering company recently laying down a track record of mega-wins in generative AI foundation model development. We have the fundamentals in place for long-term secular growth: • Recently won transformative generative AI e
Slide 5: Growing Momentum – frequency and size of mega-wins has steadily increased in 2023
01. INVESTOR HIGHLIGHTS Growing Momentum – frequency and size of mega-wins has steadily increased in 2023 Innodata Signs New Big Tech for Generative AI Development - November 2023 Innodata Lands Significant LLM Expansion with New “Big Five” Customer Announced July 18, 2023 Innodata Announces Agreeme
Slide 6: Our customers now include five of the ten largest global tech companies
03. INVESTOR HIGHLIGHTS Our customers now include five of the ten largest global tech companies building the generative AI ecosystem. The ten largest global technology companies by market capitalization: $2 Trillion+ $2 Trillion+ $1 Trillion+ $1 Trillion+ $1 Trillion+ $800 Billion+ $600 Billion+ $40
Slide 7: Our customers also include leading enterprises seeking to deploy generative and classical AI
01. INVESTOR HIGHLIGHTS Our customers also include leading enterprises seeking to deploy generative and classical AI in a range of products and operations. Audio-Based Assistants Avatar Generation Autonomous Vehicles Content Operations Contract Extraction Content Moderation Conversational AI Custome
Slide 8: Generative AI is anticipated to be the next-generation computation platform
02. GENERATIVE AI OPPORTUNITY Generative AI is anticipated to be the next-generation computation platform. We believe it will rival the internet and mobile in significance. • Big Five AI arms race now underway – billions allocated1 • McKinsey & Co. believes generative AI will potentially: o Automate
Slide 9: We're just "two steps into the [generative AI] marathon"
02. GENERATIVE AI OPPORTUNITY We’re just “two steps into the [generative AI] marathon”. 4 Opportunities (and challenges) abound • Big Five Global Tech in virtual arms race deploying billions of capex to forge path forward and discover emergent properties • Enterprises seeking to adopt to stay compet
Slide 10: Seizing opportunities and addressing challenges is requiring large-scale data engineering
02. GENERATIVE AI OPPORTUNITY Seizing opportunities and addressing challenges is requiring large-scale data engineering. High-quality, robust training data is key AI ingredient for seizing opportunities and overcoming challenges: Compute Algorithms Data Engineering 10
Slide 11: We have a full set of services and platforms to support builders and adopters of generative AI
03. HOW WE HELP We have a full set of services and platforms to support both the builders and the adopters of generative AI. Services Data Collection + Normalization LLM Integration Builders (Large Tech) Training Datasets Across Data Modalities AI Model Building and Management Synthetic Data AI Appl
Slide 12: Our data engineering capabilities span data modalities, languages, and domains
03. HOW WE HELP Our data engineering capabilities span data modalities, languages, and domains, and nine categories of tasks required by generative AI. 1- Translation 3- Logic & Semantics 4- Textual Information • Source Language to Target Language • Coherence Classification • Coreference Resolution
Slide 13: Our data engineering capabilities (continued)
03. HOW WE HELP Our data engineering capabilities span data modalities, languages, and domains, and nine categories of tasks required by generative AI. ...continued, from last page 5- Document information 6- Creation 7- Editing 8- Dialog 9- Programming • Section Classification • Paper Review • Gramm
Slide 14: Our training data results in better performance across key LLM objectives
03. HOW WE HELP Our training data results in better performance across key LLM objectives and assessments. Key Objectives Response ID: 8903X • Adapting to domains or contexts Assessed by: D.Rue • Improved model performance Date: 5.26.23 • Model behavior control • Data efficiency • Bias mitigation •