Saturday, August 02, 2025

“Misinformation by Machine: The Rise of LLM Grooming in AI Systems”

LibCognizance Disinformation , LLM Grooming

Artificial intelligence has now become a part of the lives of researchers, writers, officials and general people, impacting their decisions and planning. But what if these LLM models, such as ChatGPT, Claude, Gemini, etc., are being massively fed with wrong information and leading the users in a direction not meant for them?

#ResponsibleAI #LLMGrooming #AITrust #misinformation #Disinformation

The term "LLM grooming" was brought to light by The American Sunlight Project (ASP) [1]. In a report, the term was introduced, which describes how disinformation is being propagated by AI. These algorithmic machines are being trained massively behind the curtain to change the perception of their users about some political propaganda. Researchers from the ASP identified and investigated a network of pro-Russia propaganda websites named the Pravda network. A key finding of their research was the belief that the Pravda network may have been custom-built to flood large language models (LLMs) with pro-Russia content.

In a special reality check report released by NewsGuard [2] recently in March 2025, it has been confirmed that a well-funded, Moscow-based "news" network, named Pravda (Russian for "truth"), has been deliberately infecting Western artificial intelligence (AI) tools with Russian propaganda. They are not directly influencing human readers; they are manipulating around 10 leading generative AI models to achieve their disinformation goals by saturating search results and web crawlers with pro-Kremlin falsehoods in a significant percentage of their responses.

The Atlantic Council's DFRLab also confirmed that major AI models have cited Pravda network content.

How will you calculate the impact factor of a journal?

LibCognizance Citation metrics , Clarivate Analytics

Impact factor calculation explained by lib cognizance

Spiral of Scientific Method

LibCognizance Nadir , Spiral of Scientific Method

sp;

How to Cite ChatGPT in APA, MLA, and Chicago Styles: A Complete Guide

LibCognizance #AcademicWriting , #ChatGPT #CitingAI #CitationStyles #APAStyle #MLAStyle #ChicagoStyle

As AI tools like ChatGPT become more prevalent in academic and research writings, proper citation has become essential. This guide covers how to cite ChatGPT and other AI-generated content in APA, MLA, and Chicago styles, with examples and best practices for ethical use in scholarly work. However, the government policies, educational institutes, and style guides across the world are still working on the guidelines. This article presents the updated information to date.

Why Proper Citation of AI Matters

When using ChatGPT or similar AI tools in your research, proper citation serves two crucial purposes:

Academic integrity: Academic integrity, according to the ICAI, involves a commitment to honesty, trust, fairness, respect, responsibility, and courage. These fundamental values can be applied to many circumstances in the learning environment, including how to improve our academic writing skills to avoid plagiarism. Learning how to use AI properly, including how to cite ChatGPT and other tools, is important to pursuing and achieving academic integrity. Academic integrity acknowledges the use of AI-generated content.
Transparency: Allows readers to understand how AI contributed to your work. Citing your sources is an essential part of ethical and professional writing.

Important: Always check your institution's or publisher's guidelines before using AI tools in academic work, as policies vary widely.

General Principles for Citing AI Content

Describe how you used the AI tool in your methodology or introduction
Include the exact prompts you used
Credit the creator of the AI model (e.g., OpenAI for ChatGPT)
Never present AI-generated text as your own original writing
Verify all facts and sources provided by AI tools

How to Cite ChatGPT in Different Citation Styles

APA Style (7th Edition)

The American Psychological Association provides specific guidelines for citing ChatGPT and similar AI tools:

Element	Format	Example
Reference List Entry	Author. (Year). Title (Version) [Descriptor]. URL	OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat
In-text Citation	(Author, Year)	(OpenAI, 2023)

Example in context:
When prompted with "Explain the left brain-right brain divide," ChatGPT responded that "the notion that people can be characterized as 'left-brained' or 'right-brained' is considered to be an oversimplification" (OpenAI, 2023).

MLA Style (9th Edition)

The Modern Language Association recommends this format for citing ChatGPT:

Element	Format	Example
Works Cited Entry	"Prompt text" prompt. ChatGPT, Version, OpenAI, Date, URL	"Examples of cognitive biases" prompt. ChatGPT, 13 Feb. version, OpenAI, 16 Feb. 2023, chat.openai.com
In-text Citation	("Shortened prompt")	("Examples of cognitive")

Chicago Style (17th Edition)

Chicago style treats ChatGPT content as personal communication:

Element	Format	Example
Footnote	1. Text generated by ChatGPT, Date, OpenAI, URL	1. Text generated by ChatGPT, March 31, 2023, OpenAI, https://chat.openai.com
Subsequent Citations	2. ChatGPT	2. ChatGPT

Best Practices for Using ChatGPT in Academic Work

When to Cite ChatGPT

When quoting or paraphrasing AI-generated text
When using AI for editing or translation
When AI contributes significantly to your research process
If you’re using ChatGPT responses as a primary source

When Not to Rely on ChatGPT

As a source for factual information (it may "hallucinate" false details)
For references or citations (it often creates fake sources)
To write complete papers or sections (may be considered plagiarism)

Warning: ChatGPT's knowledge is current only until its training cutoff date (e.g., January 2022 for ChatGPT-3.5). It cannot access or verify real-time information.

Special Cases

Including Long ChatGPT Responses

For extensive AI-generated content, APA recommends:

Include a brief excerpt in your main text with citation
Place the full transcript in an appendix
Reference the appendix in your citation

When asked about neural plasticity, ChatGPT provided a detailed explanation including that "the brain's ability to reorganize itself is most pronounced in childhood but continues throughout life" (OpenAI, 2023; see Appendix A for full transcript).

Citing Other AI Tools

The same principles apply to tools like Google Bard or Bing AI. Adapt the citation format with the appropriate:

Company name (e.g., Google for Bard)
Tool name
Version information
Access URL

There are many questions over the use of ChatGPT in academic writings like, "Should students be allowed to use it? What guidelines should instructors create for students using AI? Does using AI-generated text constitute plagiarism? Should authors who use ChatGPT credit ChatGPT? What are the ? These questions are still under debate and we haven't reached on the final conclusions. Researchers, editors, instructors, and others are actively debating and creating parameters and guidelines.

References:

1. ChatGPT Citations | Formats & Examples (scribbr.com)

2. How to cite ChatGPT (apa.org)

3. How to Cite ChatGPT Sources | APA, MLA, Chicago, Vancouver - Wordvice

4. When and how to cite ChatGPT and AI in MLA / APA formats (turnitin.com)

5. How to Cite ChatGPT: A Complete Guide (linkedin.com)

6. What is Academic Integrity?

7. Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence

India's initiatives under IndiaAI Mission

LibCognizance AI , AI Compute Portal

The IndiaAI Mission: an MEITY initiative, launched by the Government of India in March 2024. Cabinet has approved over Rs.10,300 Crore for IndiaAI Mission which aims to build a comprehensive ecosystem that fosters AI innovation by democratizing computing access, enhancing data quality, developing indigenous AI capabilities, attracting top AI talent, enabling industry collaboration, providing startup risk capital, ensuring socially impactful AI projects, and promoting ethical AI. This mission drives responsible and inclusive growth of India's AI ecosystem through following seen pillars.

AIKosha: A Big Leap Toward India’s AI Mission

LibCognizance AI Compute Portal , AI Mission in India

What is AIKosha?

AIKosha is a platform that provides repository of datasets, models and use cases to enable AI innovation. It is launched by India's Ministry of Electronics and Information Technology (MeitY) for its ambitious AI Mission on March 6, 2025. The IndiaAI Mission is structured around seven core pillars, focusing on democratizing AI access, enhancing data quality, and ensuring ethical AI practices.

AIKosha is a large corpus of datasets that is built in India for growth of AI in country. Through access of curated datasets, advanced tools, and a collaborative space people can turn AI ideas into impactful solutions. It also features AI sandbox capabilities through an integrated development environment along with tools and tutorials. Speaking at the Raisina Dialogue 2025 in New Delhi, Union Minister for Electronics and IT, Ashwini Vaishnaw informed that the government has also set up a common compute structure called IndiaAI Compute Portal. He said that the MoU between the India AI Mission and the Indian Parliament was signed as Parliament has a vast dataset in multiple languages accumulated over a long period. To take the lead in AI by 2047, minister emphasized that India must have their own LLMs, continuously invest in new technologies and ensure that universities must upgrade their curricula to bring out more experts in the field.

After meeting with the founder of the Bill and Melinda Gates Foundation, Bill Gates in New Delhi. Mr. Vaishnaw also said that the Memorandum of Understanding (MoU) will be signed soon between IndiaAI mission and the Bill and Melinda Gates Foundation emphasizing that the AI solutions for better crops, stronger healthcare, smarter education and climate resilience are need of the present time.

Key takeaways?

Unified AI Resource Platform

Comprehensive Repository: AIKosha offers access to over 300 non-personal datasets and 80+ AI models, serving as a central hub for AI development. These datasets include those from the MeitY’s Digital India Bhashini division, with one dataset containing 1,684 hours of labelled speech data across 12 Indian languages, which could help develop tools for automatic speech recognition. Private companies such as Sarvam AI and Ola Krutrim have also uploaded their models to AIKosha.

Integrated Development Environment: The platform includes an AI sandbox with tools, tutorials, and secure APIs, facilitating seamless AI model training and deployment. According to Union electronics and IT minister Ashwini Vaishnaw Startups, researchers, application developers and others can now access 14,000 GPUs (graphics processing units) on the IndiaAI Compute Portal, with 4,000 more in the pipeline.

A GPU is a specialized chip built to handle massive numbers of operations in parallel. Training AI involves huge amounts of data and computation. GPUs can process many calculations in parallel, which speeds up training massively compared to CPUs.

Ethical and Secure Data Handling

Non-Monetized Data Access: The government has clarified that there are no plans to monetize the non-personal data on AIKosha, ensuring free and open access for users. Users can download the datasets by first signing up for DigiLocker using their mobile number, while organisations must register via the ministry of electronics and information technology’s (MeitY) Entity Locker using Aadhaar. This approach aims to ensure platform security and the appropriateness of the datasets.

Data Privacy Compliance: AIKosha adheres to India's data protection laws, including the Information Technology Act, 2000, and the Digital Personal Data Protection Act, ensuring responsible data usage.

Support for Indigenous AI Development

Foundation for India's AI Models: AIKosha is instrumental in developing India's own foundational AI models, including large language models tailored to the country's diverse needs.

Collaboration with Government Departments: Datasets from various ministries, such as agriculture and weather forecasting, are integrated into the platform, enriching the data pool for AI applications.

Part of a Broader AI Ecosystem

Complementary Initiatives: AIKosha is part of the larger IndiaAI Mission, which includes the AI Compute Portal providing access to over 14,000 GPUs at subsidized rates, and programs like the IndiaAI Startups Global Acceleration Program and AI Competency Framework for public sector officials.

Focus on Skill Development: Initiatives like the India AI Future Skills, the AI Competency Framework for Public Sector Officials, iGOT-AI Mission Karmayogi, an AI-powered personalized learning platform, and the establishment of AI Data Labs in Tier II and III cities aim to build a robust AI talent pool across the country.

AIKosha is a significant step in India's AI journey, providing a secure platform for AI innovation and development. The launch of AIKosha, the AI Compute Portal, and other strategic AI initiatives marks a transformative moment for India’s digital future, reinforcing its status as a pioneer in AI innovation and adoption.

References:

1. AIKosh (indiaai.gov.in)

2. https://pib.gov.in/PressReleasePage.aspx?PRID=2108961

3. https://www.newsonair.gov.in/large-public-datasets-crucial-for-growth-of-ai-in-india-union-minister-ashwini-vaishnaw/

4. https://www.newsonair.gov.in/health-minister-j-p-nadda-bill-gates-discuss-shared-commitment-to-ensure-affordable-health-care-for-all/

5. https://www.youtube.com/watch?v=2IG5VFokosA

6. https://www.facebook.com/meityindia/videos/aikosha-a-platform-that-provides-repository-of-datasets-models-and-use-cases-to-/4053665464955746/

7. https://the-captable.com/2025/02/deepseek-ai-kosh-india-meity-public-dataset/

8. No plans to monetise non-personal data on AIKosha: MeitY informs Rajya Sabha | Latest News India - Hindustan Times

9. https://pib.gov.in/PressReleasePage.aspx?PRID=2108961

10. IndiaAI Mission Turns One; Strengthens AI Ecosystem with AIKosha, Compute Portal and Key Innovations - Elets eGov (eletsonline.com)

11. https://egov.eletsonline.com/2025/03/indiaai-mission-turns-one-strengthens-ai-ecosystem-with-aikosha-compute-portal-and-key-innovations/?utm_source=chatgpt.com

12. https://www.livemint.com/technology/ai-kosh-platorm-data-sets-ai-training-models-gpu-access-live-chatgpt-nvidia-amd-yotta-openai-chatgpt-intel-aws-11741270366709.html

13. https://futuretech.media/india-takes-a-major-step-in-ai-development-with-ai-kosh-and-gpu-initiatives/

14. https://pib.gov.in/PressReleasePage.aspx?PRID=2108961

Preservation Techniques

LibCognizance Audio-visual Preservation , Digital Preservation

Preservation techniques vary based on the type of material and the risks involved (e.g., data corruption, paper deterioration, or film degradation). Digital methods like migration, emulation, and cloud storage are essential for modern archives, while traditional techniques like microfilming, deacidification, and encapsulation continue to protect physical records.

Persistent Identifiers (PIDs) and thier role in supporting scholarly communication

LibCognizance DOI , ISNI

Terms like Digital Object Identifier (DOI), OCID, International Standard Name Identifier (ISNI) are common in academic community. These all are Persistent Identifiers (PIDs).

What is Persistent identifier (PID)?

A PID (Persistent Identifier) helps to identify and locate an entity regardless of its hosting or publication location, ensuring its clear and lasting identification. PIDs play a crucial role in the research ecosystem by connecting researchers and their research outputs to the underlying data and related metadata.

Persistent Identifier includes two words, Persistent refers to anything which is long lasting, unbreakable and reliable. An identifier is a label which gives a unique name to an entity: a person, place, or thing.

As described in The Digital Preservation Handbook "A persistent identifier is a long-lasting reference to a digital resource. Typically it has two components: a unique identifier; and a service that locates the resource over time even when it's location changes. The first helps to ensure the provenance of a digital resource ( that it is what it purports to be), whilst the second will ensure that the identifier resolves to the correct current location."

The examples of PIDs include

Digital Object Identifier (DOI): It is a persistent identifiers for things or entities such as journal articles, books, and datasets. Crossref and DataCite are the main organizations assigning DOIs for these purposes in scholarly communication.

OCID: a free, unique, persistent identifier (PID) for individuals. An ORCID iD is an example of a persistent identifier for a person. ORCID works closely with Crossref, DataCite and many other PID organizations to build trusted connections between ORCID iDs and other identifiers.

International Standard Name Identifier (ISNI): This identifier provides information about the institution where a researcher worked while the research was undertaken.

Importance of PIDs in the scholarly system

Discoverability: PIDs such as DOIs, ORCID iDs, RRIDs, ROR IDs, and Funder IDs make data more easily discoverable by providing unique, permanent identifiers.

Accessibility: PIDs link research outputs to their underlying data and associated metadata, making it easier to discover and access research data.

Interoperability: Incorporating PIDs in research outputs ensures that data follows established standards, making it more interoperable with existing and future systems.

Reusability: PIDs facilitate the reuse of research data or protocols by enabling researchers to easily cite and credit the sources of their data and protocols.

Machine-Actionable Data: PIDs enable data to be processed and understood by machines or software, enhancing the efficiency of data and metadata processing.

Reproducibility and Transparency: PIDs play a critical role in ensuring the reproducibility and transparency of research data by enabling researchers to uniquely identify and cite their research resources.

Integration of Data: PIDs facilitate the integration of data from multiple sources, enabling researchers to make new discoveries that would not be possible without PIDs.

FAIR Data Principles: By incorporating PIDs in their research outputs, researchers contribute to making their data more Findable, Accessible, Interoperable, and Reusable (FAIR data principles) as required by many funders and publishers.

Open Data Ecosystem: PIDs support the open data ecosystem by ensuring the unique identification, citation, and linking of research outputs to their underlying data and associated metadata.

DataCite Commons and power of PID

DataCite Commons was developed as part of the EC-funded project Frey. The users of the DataCite Commons will have easier access to information about the use of their DOIs and can discover and track connections between their DOIs and other entities and also shows the connections between content with DOIs and people, research organizations, and funders that are together called the PID Graph of scholarly resources identified via persistent identifiers (PIDs) and connected in standard ways.

Recent advances in (PIDs) and their application in scholarly communication

Creating an ANSI/NISO standard to enhance utility of PIDs in scholarly system

Recently in a report of the Open Research Funders Group “Developing a US National PID Strategy” in March 2024. It highlighted that a strategy is required to build support for PIDs, increase their adoption, and help stakeholders incorporate them into workflows and systems more easily. Based on the principles addressed in the report while also further developing other elements, this Working Group will create a standard for advancing PIDs and open scholarship.

Finally, Research Data Alliance-United States (RDA-US) has collaborated with the National Information Standards Organization (NISO) to develop a US national PID strategy. This initiative aims to create an ANSI/NISO standard. The Standard will guide the adoption and integration of PIDs in research workflows. By doing so, it seeks to build support for PIDs, streamline their implementation, and enhance their utility across the scholarly ecosystem.

RDA-US will contribute expertise in PID implementation and community engagement, while NISO will oversee the Working Group’s operations and coordination. Leaders from both organizations express confidence that this initiative will significantly strengthen the US research infrastructure by providing clear guidance on PID adoption.

This collaboration underscores the growing recognition of PIDs as critical tools for ensuring the integrity, accessibility, and interoperability of research outputs in an increasingly digital and interconnected world.

The DOI for Scholarly Publishing: winner of the Rosenblum Award for Scholarly Publishing Impact

NISO in association with The Association of Learned and Professional Society Publishers (ALPSP), the Association of University Presses (AUPresses), the Society for Scholarly Publishing (SSP), and the International Association of Scientific, Technical & Medical Publishers (STM) has recently in February of this year announced the Rosenblum Award for Scholarly Publishing Impact. Named in honor of Bruce Rosenblum, the award celebrates innovations that have transformed the scholarly publishing ecosystem, focusing on technologies, standards, or practices that have become indispensable to its operation, and its inaugural winner is the DOI for Scholarly Publishing. Bruce Rosenblum was known for his expertise in developing Document Type Definitions (DTDs) and championing XML standards. He played a critical role in the development of the JATS and STS standards and advocated for persistent identifiers, semantic tagging, high-quality metadata, and industry standards.

Since its adoption by Crossref, the DOI for Scholarly Publishing has been critical for ensuring research objects are discoverable, even if web structures change or content moves. DOI metadata facilitates other information management systems such as holdings and appropriate-copy resolution via related standards like OpenURL. DOI metadata facilitates other information management systems such as holdings and appropriate-copy resolution via related standards like OpenURL.

The initiative involved collaboration by five sponsoring organizations: NISO, ALPSP, AUPresses, SSP, and STM. The Award Governance Committee is made up of leaders from these organizations, and there are representatives forming the Award Planning and Piloting Committee.

References:

Open source publishing platforms for scholarly research

LibCognizance Book publishing , Journal Publishing Platform

Now, we have come up with a list of free, open-source software to disseminate research and manage the entire scholarly publishing workflow, from submission to indexing, in case of books, journals, and preprints. These are the publishing software by Public Knowledge Project (PKP). From the beginning, PKP has been developing publishing platforms, such as OJS, OMP, and OPS, based on the principles and licensing of free and open-source software (FOSS). In its effort to support the publishing of open access journals, books, and preprints, PKP is an integral part of the scholarly publishing ecosystem, offering infrastructure that is as open as the science it aims to support.

For Journals	Open Journal Systems (OJS) is the world’s most widely used journal management and publishing software. Manage your entire researcher-to-reader workflow for submission, peer review, and production from one place. Download OJS See Demo See Showcase
For Books	Open Monograph Press (OMP) is an end-to-end solution for publishing books with full metadata. Publish your monographs and edited volumes with full metadata for worldwide dissemination and discovery. Download OMP See Demo See Showcase
For Preprints	Open Preprint Systems (OPS) provides everything needed to run a fully-featured preprint server for researchers. Accelerate research by allowing researchers to upload datasets, revise papers, and link preprints to the final published work. Download OPS See Demo See Showcase
Hosting Services	PKP Publishing Services can host your OJS, OMP, and OPS installation on professionally maintained and secured servers with guaranteed uptime. Hosting Plans

Puzzle 4

LibCognizance library science puzzles , Puzzle

👉 Answers

3. a the company that developed and released Chatgpt.
5. a cloud-based interlibrary loan (ILL) management system by OCLC.
6. an input that a user feeds to an AI system in order to get a desired result or output.
9. The company that developed CiteScore.

1. The Open Data Format meets these Guiding Principles for scientific data management and stewardship.
2. It is also called green open access model.
4. Headquarter of the National Digital Library of India
5. It is a basic unit of text that an LLM uses to understand and generate language.

AGORA: The Ultimate Hub for AI Governance Documents

LibCognizance AGORA , AI

Artificial intelligence (AI) is fascinating everyone and influencing almost every aspect of life, directly and indirectly. AI tools are widely used across various domains. From professional workflows to everyday personal tasks, AI is becoming an indispensable tool in enhancing productivity and creativity. Tools like chatbots, virtual assistants, and automated data processing software help professionals streamline their workflows, improve efficiency, and save time. Students benefit from AI-driven platforms that assist in writing assignments, understanding complex topics, and learning new skills through interactive tools. Researchers and scientists rely on AI for data management, conducting analyses, and even drafting and reviewing academic papers. AI accelerates scientific discovery by processing large datasets quickly and uncovering patterns that humans might miss.