SUMMARY & CALL FOR INPUTS ON THE CONSULTATION PAPER ON LEVERAGING ARTIFICIAL INTELLIGENCE & BIG DATA IN THE TELECOMMUNICATIONS SECTOR

Terms of use

Terms of Use

The use of this site and the content contained therein is governed by the Terms of Use. When you use this site you acknowledge that you have read the Terms of Use and that you accept and will be bound by the terms hereof and such terms as may be modified from time to time.

All text, graphics, audio, design and other works on the site are the copyrighted works of nasscom unless otherwise indicated. All rights reserved.
Content on the site is for personal use only and may be downloaded provided the material is kept intact and there is no violation of the copyrights, trademarks, and other proprietary rights. Any alteration of the material or use of the material contained in the site for any other purpose is a violation of the copyright of nasscom and / or its affiliates or associates or of its third-party information providers. This material cannot be copied, reproduced, republished, uploaded, posted, transmitted or distributed in any way for non-personal use without obtaining the prior permission from nasscom.
The nasscom Members login is for the reference of only registered nasscom Member Companies.
nasscom reserves the right to modify the terms of use of any service without any liability. nasscom reserves the right to take all measures necessary to prevent access to any service or termination of service if the terms of use are not complied with or are contravened or there is any violation of copyright, trademark or other proprietary right.
From time to time nasscom may supplement these terms of use with additional terms pertaining to specific content (additional terms). Such additional terms are hereby incorporated by reference into these Terms of Use.

Disclaimer

The Company information provided on the nasscom web site is as per data collected by companies. nasscom is not liable on the authenticity of such data.
nasscom has exercised due diligence in checking the correctness and authenticity of the information contained in the site, but nasscom or any of its affiliates or associates or employees shall not be in any way responsible for any loss or damage that may arise to any person from any inadvertent error in the information contained in this site. The information from or through this site is provided "as is" and all warranties express or implied of any kind, regarding any matter pertaining to any service or channel, including without limitation the implied warranties of merchantability, fitness for a particular purpose, and non-infringement are disclaimed. nasscom and its affiliates and associates shall not be liable, at any time, for any failure of performance, error, omission, interruption, deletion, defect, delay in operation or transmission, computer virus, communications line failure, theft or destruction or unauthorised access to, alteration of, or use of information contained on the site. No representations, warranties or guarantees whatsoever are made as to the accuracy, adequacy, reliability, completeness, suitability or applicability of the information to a particular situation.
nasscom or its affiliates or associates or its employees do not provide any judgments or warranty in respect of the authenticity or correctness of the content of other services or sites to which links are provided. A link to another service or site is not an endorsement of any products or services on such site or the site.
The content provided is for information purposes alone and does not substitute for specific advice whether investment, legal, taxation or otherwise. nasscom disclaims all liability for damages caused by use of content on the site.
All responsibility and liability for any damages caused by downloading of any data is disclaimed.
nasscom reserves the right to modify, suspend / cancel, or discontinue any or all sections, or service at any time without notice.

For any grievances under the Information Technology Act 2000, please get in touch with Grievance Officer, Mr. Anirban Mandal at data-query@nasscom.in.

New

See all

No notification found.

SUMMARY & CALL FOR INPUTS ON THE CONSULTATION PAPER ON LEVERAGING ARTIFICIAL INTELLIGENCE & BIG DATA IN THE TELECOMMUNICATIONS SECTOR

Sudipto

@Sudipto

September 20, 2022

Public Policy

203

Text

Description automatically generated

SUMMARY & CALL FOR INPUTS ON

THE CONSULTATION PAPER ON LEVERAGING ARTIFICIAL INTELLIGENCE & BIG DATA IN THE TELECOMMUNICATIONS SECTOR

BY THE TELECOM REGULATORY AUTHORITY OF INDIA

KEY TAKEAWAYS

The Telecom Regulatory Authority of India (TRAI) had, on August 5^th, published a Consultation Paper (CP) on 'Leveraging Artificial Intelligence and Big DaTa in Telecommunication Sector’.
The CP poses both general questions on the regulation and adoption of artificial intelligence (AI) and big data (BD) across sectors and sector-specific questions.
We provide below an overview of the CP and group the TRAI’s consultation questions into sets for the benefits of members to approach the CP and to prepare their comments.
The deadline for submitting written comments to the TRAI on the CP by October 14^th. Counter-comments, if any, may be submitted by October 28^th.
We request members to send in their written inputs to us by October 7^th, 2022.
Inputs may be sent to deepa.ojha@dsci.in and anisha@dsci.in [at DSCI] and to and sudipto@nasscom.in and varun@nasscom.in [at NASSCOM].

STRUCTURE OF THE CONSULTATION PAPER

Chapter 1 provides background and deliberates upon prior efforts of the Indian telecom regulators to drive the adoption of new technologies in the telecom sector.
Chapter 2 deliberates on the concepts of AI and BD.
Chapter 3 discusses opportunities of AI and BD in the telecom sector and highlights the telecom sector’s role in the leveraging of AI and BD in other sectors for providing access to data, offering execution environment, providing a privacy preserving architecture, and developing a federated model to learn. Risks and concerns involved in adoption of AI and BD are also discussed.
Chapter 4 discusses the key constraints in adoption of AI and BD in the telecom sector and in offering its capabilities to others.
Chapter 5 deliberates on possible ways to accelerate adoption of AI and BD.
Chapter 6 lists out issues for consultation.

OVERALL APPROACH OF THE TRAI

The starting point of the CP is to encourage the use of AI and BD in the telecom sector. It presents a number of use-cases of AI & BD in areas such as quality of service, network security and spectrum management. It then discusses how the rapid adoption of these technologies necessitates a cogent policy on how they can be employed to grow and develop the sector.
However, the CP then also notes that there are several use-cases in other sectors as well, and then notes that it is also relevant to examine how the telecom sector can enable the use of these technologies in other sectors as well. As a result, the CP also adopts a cross-sectoral lens.
While the CP does not explicitly recognise this, we find that it effectively contains discussions and poses questions under two themes:
- Cross-sectoral adoption of AI & BD: like questions on the definition of AI, the types of risks in AI implementation, the concept of having a “responsible AI framework”, privacy and security concerns connected to these technologies, the need for an AI regulator, etc.
- Sectoral adoption of AI & BD in telecom: like questions on leveraging these technologies to improve connectivity in the 5G era, build better experiences for end-users, improve spectrum management, etc.

Discussion under Theme 1: Adoption of AI and BD across sectors

The CP notes that there is no globally accepted definition of AI and hence offers multiple definitions provided by different agencies such as ETSI. The CP states that governments, regulatory agencies, international agencies, and other forums have adopted varying definitions depending on the context before them.
The CP recognizes that AI models utilise large datasets to train, evaluate, and optimise the models and deems it important to understand BD. Most of the organisations and companies characterise BD by the volume, velocity and variety of data being produced. However, the CP notes that a report by the ITU also includes veracity which can be understood as “biases, noise and abnormality in data”. The CP acknowledges that the telecommunications industry “generates and stores tremendous amount of data”.
The CP identifies possible risks and challenges associated with AI, like data biases, data poisoning, model biases, privacy, data exploitation, risk of identification and tracking, and individual profiling. The CP also deals with concepts, like, Responsible AI, Explainable AI, and Trustworthy AI. However, TRAI notes that these terms may be overlapping and may be difficult to identify and hence questions related to these themes have been posed by the regulator.
The TRAI notes that there are certain risks involved with the use of AI such as unethical use cases, insufficient learning from feedback loop, implementation errors, biased or discriminatory model outcomes, model instability or performance degradation, insufficient training, regulatory non-compliance etc.
In addition to the risks associated with the use of AI and BD in telecom sector, the CP sheds light on some of the key constraints that may hamper the use of AI and BDl, like, lack of data accessibility as most of the data is either available with the big organisations or with the government agencies. Further, there is a lack of AI specific infrastructure and research and development in the country. Hence, the CP has posed questions related to the same.
Some of the solutions offered by TRAI to address these constraints include data democratisation, referring to data sharing by people, industry, and the government. It states that there is a requirement to establish an authority or a body or an institution whose role should be to act as a manager and gatekeeper for data stored along with managing the privacy and security of data.
The CP offers ways in which privacy can be protected without hampering rights of people, like data anonymization, differential privacy, secure multi-party computation, homomorphic encryption.
The CP also initiates a discussion on the regulation of AI, by posing questions on the efficacy of the existing legal framework, and whether there is a need for a new regulator or not.

Discussion under Theme 2: Adoption of AI And BD in the Telecom Sector

The CP explains that AI advancement will work towards improving 5G systems performance and efficiency while expansion of 5G will drive distributed intelligence through connected devices. Furthermore, 5G is expected to “generate massive amounts of data from multitude of new services, and billions of IoT devices, which will enhance AI learning and model training.”
The CP thus seeks input to understand whether deployment of 5G and beyond technologies will help to accelerate adoption of AI in all the sectors. The CP has identified several opportunities of AI and BD in telecom, like, network Design, improving digital connectivity inside buildings, international use cases by telecom operators in network planning, network optimisation & configuration, energy saving and efficiency improvement, network maintenance and spectrum management.

QUESTIONS FOR CONSULTATION (GROUPED THEMATICALLY)

Questions under theme 1: Cross-sectoral adoption of AI & BD

Definitions & state of the technology

What may be the most appropriate definition of Artificial Intelligence (AI)? Please justify your response with rationale and global practices, if any. (See question 1, CP)
What type of technological advancements are happening for running the AI models on the end user devices to overcome constraints in respect of processor, memory, battery etc.? Whether special tools, programming languages, and skills are required to be developed to build such AI models? Please justify your response with rationale and suitable examples, if any. (See Question 22, CP).

Responsible AI

Do you think that a number of terminologies such as Trustworthy AI, Responsible AI, Explainable AI etc. have evolved to describe various aspects of AI but they overlap and do not have any standardised meanings? If yes, whether there is a need to define or harmonise these terms? Please justify your response with rationale and global practices, if any. (See question 4, CP)
What measures do you suggest to instil trust and confidence regarding a robust and safe AI system among customers, TSPs and other related entities/stakeholders? Whether adopting general principles such as Responsible AI and ethical principles at the time of designing and operationalising the AI models will help in developing ethical solutions and instilling trust and confidence in the users? What may be such principles and who should formulate these and how compliance can be ensured? Please justify your response with rationale and suitable examples, if any. (See question 10, CP)

Risks and concerns

Whether risks and concerns such as privacy, security, bias, unethical use of AI etc. are restricting or likely to restrict the adoption of AI? List out all such risks and concerns associated with the adoption of AI. Please justify your response with rationale and suitable examples, if any. (See question 8, CP)
What measures are suggested to be taken to address the risks and concerns listed in response to Q.8? Which are the areas where regulatory interventions may help to address these risks and concerns? Please justify your response with rationale and suitable examples, if any. (See question 9, CP)

Privacy and security

Which are the currently used privacy enhancing and privacy preserving technologies facilitating adoption of AI and BD? Are there any challenges in using these technologies? How these challenges can be addressed? Please justify your response with rationale and suitable examples, if any. (See Question 19, CP).
Which are the potential technologies likely to be available in near future to further strengthen privacy? Please justify your response with rationale and suitable examples, if any. (See Question 19, CP).
Whether the list of technologies provided in response to Q.19 are adequate to handle all the perceived risks and concerns in the AI domain? Or is there a need to develop new privacy preserving architecture? Please justify your response with rationale and suitable examples, if any. (See Question 20, CP).
Whether the next generation telecom network architectures such as AI at edge, federated learning, TinyML or their combination can offer solutions to meet both privacy as well as intelligence requirements? Please justify your response with rationale and suitable examples, if any. (See Question 21, CP).
Considering availability of new privacy preserving architectures as suggested in response to Q.19 and Q.20, what is the likelihood of emergence of new business and operational models? Whether such models will raise issues related to ownership and responsibilities? What do you suggest to address these issues? Please justify your response with rationale and suitable examples, if any. (See Question 23, CP).
Whether AI/ML developers should launch bounty programs to establish trust in the public about robustness of measures taken by them to protect privacy in their products or solutions? Whether conduction of such programs will help companies or firms to improve their products or solutions? Whether such programs should be conducted under the supervision of the government, or an institution established/assigned for this purpose? Please justify your response with rationale and suitable examples, if any. (See Question 31, CP).

Legal and regulatory

Whether there is a need of telecom/ICT sector specific or a common authority or a body or an institution to check and ensure compliance of national level and sector specific requirements for AI? If yes, what should be the composition, roles and responsibilities of such authority or body or institution? Please justify your response with rationale and suitable examples or best practices, if any. (See question 11, CP)
In response to Q.11, if yes, under which present legal framework or law such authority or body or institution can be constituted and what kind of amendments will be required in the said law? Or whether a new law to handle AI and related technologies is a better option? Please justify your response with rationale and suitable examples or best practices, if any. (See Question 12, CP)
Whether the authority or body or institution as suggested in response to Q.11 may also be entrusted with the task to manage and oversee collection, cataloguing and storage of data? Whether such authority or body or institution need to be entrusted to generate and make available synthetic data? Please justify your response with rationale and suitable examples, if any. (See Question 17, CP)
Whether the legal framework as envisaged in para 3.5.3 and Q.12 should also enable and provide for digitalisation, sharing and monetisation for effective use of the data in AI without affecting privacy and security of the data? Please justify your response with rationale and suitable examples, if any. (See Question 18, CP).
Whether experiments are required to be backed by regulatory provisions such as regulatory sandbox to protect experimenters from any violation of existing regulations? Whether participation of government entities or authorities during experimentation will help them to learn and identify changes required in the existing regulations or introducing new regulations? Please justify your response with rationale and suitable examples, if any. (See Question 28, CP).
Whether establishing a system for accreditation of AI products and solutions will help buyers to purchase such solutions or products? If yes, what should be the process of accreditation and who should be authorised or assigned with the task of accrediting such products or solutions? Please justify your response with rationale and suitable examples, if any. (See Question 35, CP)
Whether creating a framework to prepare a list of prequalified suppliers of AI products or solutions will help industry including government agencies to procure AI products or solutions? Whether there is a need to formulate a standard Code of Conduct or guidelines for AI related procurements? What should be the typical elements of such a Code of Conduct or guidelines including guidelines on trusted source and who should be tasked to formulate such a Code of Conduct or guidelines? Please justify your response with rationale and suitable examples, if any. (See Question 36, CP).

Questions under theme 2: Adoption of AI & BD in the telecom sector

Which are the applications of AI and BD already being used by the TSPs in their networks to improve Quality of Service, Traffic Management, Spectrum Management and for Security purposes? Please list out all such applications along with the level of maturity of such applications. Please specify whether they are at trial stage or pilot stage or have reached the deployment stage? Details should include type of AI models, methods to access data, and procedures to ensure quality of data. (See Question 5, CP)
What are the major challenges faced by the telecom industry, including policy and regulatory, in developing, deploying, and scaling applications of AI listed in the response to Q.5? How can such challenges be overcome? Please justify your response with rationale and suitable examples, if any. (See Question 6, CP)
In which areas of other sectors including broadcasting, existing and future capabilities of the telecom networks can be used to leverage AI and BD? Please justify your response with rationale and suitable examples if any. (See Question 7, CP)
Whether there is a need to create AI-specific infrastructure for the purpose of start-ups and enterprises in the telecom sector to develop and run AI models in an optimised manner? Whether such an infrastructure should cover various real-world scenarios such as cloud AI, edge AI and on-device AI? Please justify your response with rationale and suitable examples, if any. (See Question 25, CP)
Whether the telecom industry is required to adopt a Machine Learning Operations (MLOps) environment to develop, train, validate and store ML models? Whether there is also a need to establish a DataOps feature store to help MLOps for training purposes? What standardisation is required in terms of interoperability and compatibility for MLOps to function in a federated manner? Please justify your response with rationale and suitable examples, if any. (See Question 32, CP)

Ecosystem building

What measures are required to make data and computing infrastructure available and accessible to developers and also to make data/AI models interoperable and compatible? Please respond along with examples, best practices and explanatory notes (see Question 14, CP).
Whether there is a gap between requirement and availability of skilled AI workforce? If so, what measures are required to be taken to ensure availability of adequate skilled workforce in AI domain? Please respond along with suggestions with supporting details and best practices (see Question 15, CP).
Whether there is a need to establish experimental campuses where start-ups, innovators, and researchers can develop or demonstrate technological capabilities, innovative business and operational models? Whether participation of users at the time of design and development is also required for enhancing the chances of success of products or solutions? Whether such a setup will reduce the burden on developers and enable them to focus on their core competence areas? Please justify your response with rationale and suitable examples, if any. (see Question 27, CP)
Whether there is a need to establish experimental campuses where start-ups, innovators, and researchers can develop or demonstrate technological capabilities, innovative business and operational models? Whether participation of users at the time of design and development is also required for enhancing the chances of success of products or solutions? Whether such a setup will reduce the burden on developers and enable them to focus on their core competence areas? Please justify your response with rationale and suitable examples, if any.
Whether experiments are required to be backed by regulatory provisions such as regulatory sandbox to protect experimenters from any violation of existing regulations? Whether participation of government entities or authorities during experimentation will help them to learn and identify changes required in the existing regulations or introducing new regulations? Please justify your response with rationale and suitable examples, if any.
In response to Q.27 and Q.28, whether establishing such a campus under government patronage will enable easy accessibility of public resources such as spectrum, numbering and other resources to the researchers? Whether it would be in mutual interest of established private players as well as start-ups, innovators and enterprises to participate in such experiments? Please justify your response with rationale and suitable examples, if any. (See Question 29, CP).
Whether active participation in the international challenge programs such as ITU AI/ML 5G challenge will help India’s telecom industry in adopting AI? Whether similar programs are also required to be launched at the national level? Whether such programs will help to curate problem statements or help in enabling, creating, training and deploying AI/ML models for Indian telecom networks? What steps or measures do you suggest to encourage active participation at international level and setting up of such programs at national level? Please justify your response with rationale and suitable examples, if any. (See Question 30, CP).
Whether active participation in the international bootcamp programs such as MIT Bootcamps, Design Thinking Bootcamp by Stanford University etc. will help India’s telecom industry workforce to find international developers’ community, navigate challenges and learn from experiences of others? Whether similar programs are also required to be launched at the national level? What steps or measures do you suggest to encourage active participation at the international level and setting up of such programs at the national level? Please justify your response with rationale and suitable examples, if any. (See Question 33, CP).
Whether the courses or programs related to AI/ML currently being offered by various institutions and universities in India are adequate to meet the capacity and competence required to develop and deploy AI solutions or products in the telecom networks? If not, what additional steps or measures are suggested to fill the gap? Please justify your response with rationale and suitable examples, if any. (See Question 34, CP).
Whether there is a need to prepare and publish a compendium of guidance, toolkits and use cases related to AI and BD, to foster adoption in the telecom sector? If yes, what should be the process to prepare such a compendium and who should be assigned this task? Please justify your response with rationale and global best practices, if any. (See Question 37, CP)
Whether there is a need to establish telecom industry-academia linkages specifically for AI and BD to accelerate the development and deployment of AI products and solutions? Whether there is a need to establish Centres of Excellence (CoEs) for this purpose or it can be achieved by enhancing the role of existing TCoE? Please justify your response with rationale and global best practices, if any. (See Question 38, CP)
Whether there is a need to establish telecom industry-academia linkages specifically for AI and BD for AI related skill development? Please give the suggestions for strengthening the industry-academia linkages for identification of the skill development courses. Please justify your response with rationale and global best practices, if any. (See Question 39, CP).

Miscellaneous

Any other issue which is relevant to this subject? Please suggest with justification. (See Question 40, CP)

***

5g telecom AI ML

Disclaimer

That the contents of third-party articles/blogs published here on the website, and the interpretation of all information in the article/blogs such as data, maps, numbers, opinions etc. displayed in the article/blogs and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party article/blogs published, are provided solely as convenience; and the presence of these articles/blogs should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these articles/blogs , you do so at your own risk.

Sudipto

Call for Inputs: SEBI's consul...

Swapnil