WARNING: Your PhD Might Get DRASTICALLY Easier With This AI (Genspark Demo: Auto Drafts & Slides)

Gen Spark AI: A Deep Dive for Academia and Research

Key Concepts:

• Gen Spark: An AI tool for academic and research purposes, similar to Manis AI.
• Super Agent AI: A suite of advanced AI agents within Gen Spark.
• AI Slides: A feature that generates presentations from raw data.
• Deep Research: A tool for literature reviews and paper drafting.
• Hallucination: The generation of false or nonsensical information by an AI.
• Four-point probe: A technique for measuring the conductivity of materials.
• Resistivity: A measure of a material's resistance to the flow of electric current.
• Annealing: A heat treatment process used to alter the physical and chemical properties of a material.
• Graphical Abstract: A visual summary of a research paper.

1. Literature Review Generation

• Process: The user provides a prompt (e.g., "Write a literature review on nano composite transparent electrodes and their use"). Gen Spark creates a plan, then generates the review in real-time, showing the typing process.
• Strengths: The generated text is in-depth, well-structured, and includes bullet points. It provides information on materials used and is properly referenced.
• Weaknesses: The number of sources (18 in the example) is lower compared to other AI tools like Manis AI, which can provide 50-60+ references for scientific literature reviews.
• Output: A "spark page" is generated with easy navigation, question-answering capabilities, and source links.
• Example: The literature review on nano composite transparent electrodes included a video summary related to the topic.
• Assessment: Rated A- or B+ due to the quality of the text but the limited number of references.

2. Peer-Reviewed Paper Draft Generation

• Process: The user provides figures (images with captions) and a prompt (e.g., "Use these figures to create a story structure for a peer-reviewed paper submission").
• Functionality: Gen Spark analyzes the figures, extracts data from the captions, and proposes a paper structure. It can then generate a full paper draft.
• Strengths: The AI creates an abstract, incorporates figures into the draft (not just referring to them), and orders them logically, even if the order differs from the original labeling.
• Example: Using figures related to an unspecified topic, Gen Spark generated a complete paper draft with an abstract, introduction, results, and discussion.
• Argument: The presenter believes this is the future of research, where AI generates the first draft of a paper from data.
• Output: A downloadable PDF of the paper draft.
• Assessment: Considered a "fantastic start" and one of the best first drafts of a peer-reviewed paper generated by AI.

3. Graphical Abstract Creation

• Process: The user provides the abstract of a paper and asks Gen Spark to create a graphical abstract.
• Result: The generated graphical abstract is nonsensical, with incorrect text and imagery.
• Assessment: The graphical abstract creation is deemed "not good at all" and not suitable for academic use. Other tools like ChatGPT perform better in this area.
• Conclusion: Gen Spark is not recommended for creating graphical abstracts.

4. AI Slides (Presentation Generation from Raw Data)

• Process: The user uploads raw data files (e.g., CSV files from a four-point probe measurement) and provides a prompt (e.g., "Can you turn these four-point probe data files into a presentation?").
• Functionality: Gen Spark converts file formats, summarizes data, generates figures, and creates presentation slides.
• Example: Using two raw data files from four-point probe measurements of carbon nanotube materials, Gen Spark generated a presentation with an introduction, methodology, unannealed sample comparison, and key findings.
• Strengths: The AI can infer the purpose of the data, calculate key values (e.g., resistivity), and present the information in a structured and visually appealing way. It can also generate an introduction and methodology section.
• Output: The presentation can be viewed as HTML code with animations and can be saved as a PowerPoint file (though formatting may be imperfect in the current beta version).
• Data: The AI correctly identified a 14.5% narrow gap after annealing based on the provided data.
• Argument: This feature allows academics to produce presentations from raw data in a one-step process.
• Assessment: Considered "pretty bloody good" and a valuable tool for academics.

Synthesis/Conclusion

Gen Spark is a promising AI tool for academia and research, with strengths in generating literature reviews, drafting peer-reviewed papers from figures, and creating presentations from raw data. While the graphical abstract creation is currently weak, the AI Slides feature stands out as a unique and powerful tool for researchers. The presenter believes that AI-assisted paper drafting is the future of research, despite potential resistance from journals and universities.