1
I NTRODUCTION Insights as Stories Ganes Kesari Co-founder & Head 100+ Clients of Analytics “Simplify Data Science for all” Help apply & adopt Analytics Our data science platform, Gramex is now open- sourced! 2
Smart Contract Ris isk k Id Identif ification wit ith AI 3
AI VS L AWYERS : T HE U LTIMATE SHOWDOWN 4 https://www.lawgeex.com/resources/AIvsLawyer/ https://hackernoon.com/20-top-lawyers-were-beaten-by-legal-ai-here-are-their-surprising-responses-5dafdf25554d
C HALLENGES WITH C ONTRACT R EVIEW Deliberate Voluminous Tedious Verbosity Error-prone Lower cognition 5
H OW CAN T ECHNOLOGY H ELP ? 6 Images: Gmail; Grammarly; Skype
H OW DO THESE W ORK ? Machine Learning “Programs that solve “Programs that learn vs the problem” to solve the problem” 7
H OW DO M ACHINES LEARN ? Known Known Input Outcome Machine learning how to do the job Desired New Input Outcome …210.5 … $204.5, 207.4, 207.1, 205.2, AAPL 8
A NALYTICS IS MORE THAN M AGIC WITH N UMBERS ”What a nice sunny …day? 9 Icons: Watercolor vector by starline on freepik; flaticon on freepik; all-free-download.com
W HY D EEP L EARNING ? Traditional Machine Learning Identify features Person Input Output to teach model Name Deep Learning Model automatically identifies Person Input Output 10 features to learn Name 10 https://www.cs.toronto.edu/~ranzato/publications/taigman_cvpr14.pdf
N OT VERY DIFFERENT FROM HOW WE LEARN Training ..Versatile Detection! 11
AI AROUND US - R EADING T EXT 12
AI AROUND US – T RANSLATING T EXT 13 Google Translate
AI AROUND US – G ENERATING T EXT 14 OpenAI: https://openai.com/blog/better-language-models/
O UR W ORK ON C ONTRACT R ISK I DENTIFICATION ✔︐ Categorize ✔︐ Assess Risk ✔︐ Simplify Reviews 15
P REPARING DATA TO T RAIN THE AI Data: Categories: Agreement Liability-Indemnity 100+ Documents Confidentiality Others 6000+ Clauses Dispute Resolution Payment Terms IP Rights Scope Terms And Conditions Sub Contract Warranty 16 Icon by Smashicons on Freepik
P REPARING DATA TO T RAIN THE AI Term. This Agreement shall be in force for a period of two (2) years from the Effective Date. This Agreement may be terminated by either Party at any time upon thirty (30) days written notice to the other Party. Provided however that the termination of this Agreement shall not relieve any Party of its obligations with respect to Confidential Information disclosed under this Agreement for a period of five (5) years after the expiry/termination of this Agreement. Categories: Criticality: Agreement Liability-Indemnity Non-Critical ✔︐ Critical Confidentiality Others Dispute Resolution Payment Terms IP Rights Scope ✔︐ Terms And Conditions Sub Contract Warranty 17
B UILDING THE M ODEL Data Preparation Modeling Gathering ULM Fit Language model Built a Classifier – Criticality & Structuring Labeling Clause type Start End Transfer Learning User Workflow Pretrain on WikiText Interactive UI Train on Legal text Human augmentation 18
L ANGUAGE M ODEL P RIMED ON OUR C ONTRACTS Gramener Technology will be…. Gramener Technology will be… the sole and exclusive owner of Gramener ’s Confidential 1. Information and the Project Plan is for the sole purpose of exploring the possibility of Data Analytics , Visualization and Consulting Services . xxbos This Master Services Agreement ( “ Agreement ” ) is made on 16th February , 2015 ( “ Effective Date ” ) xxbos The Software Services and Maintenance Services Guidelines will be provided by the Client. Gramener Technology will be… a company incorporated under the provisions of the Companies 2. Act , 1956 or having its registered office at Plot No.9 / 2 , 2nd Floor , Sy . No.64 , HUDA Techno Enclave , Phase – II , Madhapur , Hyderabad - 500081 , Telangana , India ( “ Gramener ” ) , ( " Vendor " ) and the place of supply to Client ( hereinafter referred to as " the Receiving Party " ). 19
D EMO 20
T AKEAWAYS Factor Acquire right Structure & Plan for business Data Clean Data labelling dynamics Plan for Sensitize on Build into Augment with refresh accuracy workflow human inputs 21
H UMAN + AI “ Machines will partner and cooperate with humans rather than become mankind ’ s biggest enemy. - Jack Ma, Founder of Alibaba 22
Thank You! Presentation deck available at gkesari.com/LawAI gramener.com @kesaritweets @kesari 23
Recommend
More recommend
