The AI Conference 2017
as the organisers put it
The premier AI conference. Sept 29–Oct 1, 2026 at Pier 48, San Francisco. 5,500+ builders, researchers & leaders. 120+ speakers. 5 tracks covering AGI, LLMs, agentic AI, infrastructure & applied AI.
who it's for
conversational AI engineers, Alexa developers, applied machine learning engineers, machine learning infrastructure engineers, startup founders and executives, AI safety researchers, product managers for AI
topics
AI safety hard boundaries; VC due diligence for AI startups; in-car gaze tracking and vision; end-to-end speech recognition at scale; hate speech moderation pipelines; machine reading comprehension systems; intent extraction and clustering; semantic parsing for backend APIs; materials data mapping with physics-aware ML; Bayesian hyperparameter optimization API; far-field wake-word and beamforming systems
| title | author | length |
|---|---|---|
| Dealing with AI as a Dystopian Threat to Humanity | Igor Markov | 19m |
Argues that runaway AI is realistic only if it controls physical resources and offers four concrete defensive constraints: hard boundaries between intelligence… | ||
| A VC Perspective on AI | Malika Cantor | 21m |
Comet Labs runs a $10M early-stage fund plus an in-house lab that secures proprietary data access from corporations, runs pilots with Fortune 500s, and helps… | ||
| Machine Learning Inside the Car | Shalini De Mello | 28m |
Data show 90% of car crashes involve driver distraction, so NVIDIA designs in-car vision systems to keep the human in the loop until full autonomy arrives. The… | ||
| The Story of End to End Models in Deep Learning | Sanjeev Satheesh | 27m |
Baidu’s Deep Speech work claims that sufficiently large end-to-end nets beat hand‑engineered pipelines by directly mapping audio to text using models trained… | ||
| Emotional Trauma and Machine Learning | Caroline Sinders | 32m |
Work combining IBM Watson chatbot design, five years on Wikimedia’s anti-harassment team and interviews with over 40 Gamergate victims leads to a concrete… | ||
| Building Literate Machines | Rahul Mehrotra | 34m |
Maluuba (now part of Microsoft) builds neural systems that read web pages and answer questions in real time; a live demo shows a Chrome extension parsing… | ||
| How Might Artificial Intelligence Come About? | David Brin | 33m |
Maps six general approaches to creating minds — six general pathways — naming logic/knowledge systems, neural nets and learning, emergent network effects… | ||
| Why Chatbots Fail, And How To Fix Them | Artemy Malkov | 13m |
Analyzes thousands of chatbot logs and shows 10–30% misunderstanding per user utterance that multiplies across turns, producing retention rates 3–5× lower than… | ||
| Conversational Bots: Panel Discussion | Artemy Malkov, Bear Douglas, Jayesh Govindarajan | 37m |
Panelists debate why chatbots stumble and give concrete tactics: training data and the long tail force firms to constrain the domain or suffer noisy models… | ||
| Startup Showcase | John Melas-Kyriazi, Bryce Meredig, Will Murphy, Scott Clark | 57m |
Platform that maps materials data from PDFs and handbooks into searchable structured records and runs physics-aware ML and simulation to cut R&D cycles 2–5x… | ||
| Conversational AI in Amazon Alexa | Ashwin Ram | 42m |
Overview of how Alexa actually works: far‑field microphone arrays plus on‑device beamforming detect the wake word and send audio to cloud ASR and NLU that must… | ||
| nothing matches | ||