Google Gemini vs Bluedot

7

October

2025

No ratings yet.

In recent months, I have been experimenting with various AI tools to enhance the efficiency of my work, particularly tools that make mundane tasks (that generally take a fair amount of my time) at work easier. More specifically, I was trying out tools that record meetings, summarize their content, and automatically create action points. After experimenting with various tools, Bluedot and Google Gemini stood out from the meeting assistants. Both tools have very similar functionalities and can be embedded into the search browser. However,  what I found out is that their effectiveness is very different.

From the very beginning, Bluedot was proving to be much more accurate and effective than Google Gemini. It did not simply transcribe the meetings; it appeared to understand the point of them. Even if there were multiple participants or a lot of jargon (such as technical or business vocabulary), Bluedot was still able to refine concepts and translate them into precise transcripts, notes and recordings. The interface is user-friendly, and the AI-generated notes are organized so well that if I had to, I could directly share them with the participants. It creates complete sentences, logical sequences, and nicely ordered lists of action items. Bluedot also excels at separating speakers and retaining continuity across relevant points of discussion.

In contrast, Google Gemini seemed promising with the fact that it is embedded in my Google Workspace, but it was not the most consistent tool. While it performed fairly well with simple discussions, it frequently misheard or misinterpreted critical signals or phrases when participants spoke fast or when strong accents were present. As a consequence, the generated notes missed the primary intention of the meeting or summarized irrelevant references rather than the important ones. Moreover, Gemini’s summaries were far more generic than the Bluedots’, as they were missing a level of nuance and prioritized perspective that made Bluedots’ notes really valuable. However, Google Gemini does have an advanced function where it sends the meeting notes and action points to participants right away, which Bluedot does not have.

My experiments revealed that accuracy in language understanding is a critical challenge for generative AI. Even the most sophisticated large language models can fail when it involves context, accents, or technical vocabularies. For generative AI to be useful, tools such as Gemini could implement domain-adaptive fine-tuning (i.e., learning from a user’s recurring topics or project vocabulary) and include real-time feedback loops to provide corrections on summaries to adapt to the learning process.

Ultimately, my experience confirmed that while generative AI tools can be very compelling, their value is dependent on comprehension and trust, something Bluedot certainly does much better than Gemini in the context of productivity in meetings. I am interested to hear, have you tried any AI tools that enhance the productivity of your work?

Please rate this

Are AI Influencers the Future of Online Fame?

12

September

2025

No ratings yet.

AI influencers are no longer novelty items and are now a real part of the creator economy. Lil’ Miquela, a created character that first appeared in 2016, has walked red carpets, appeared in campaigns for companies such as Prada and Calvin Klein, recorded songs, and gained millions of followers (Lil Miquela, 2025). Miquela’s never tried to hide the fact that she is artificial, but has been able to build trust with followers who interact with her as if she were real (Economic Times­, 2025).


Now let’s talk about Mia Zelu, a newer AI influencer who drew lots of attention when her photorealistic impressions of images at Wimbledon went viral (Mia Zelu, 2025). A fair amount of her fan base thought she was at Wimbledon, even though her profile did have a slight, tucked disclaimer that stated she was a “digital creator & influencer AI.” The potential of plausibility with some disclaimers highlights the opportunity and risk that exists with artificial figures; they may be highly convincing and yet also possibly mistaken at an non-aware interval to be a real person (Independent, 2025).


The EU AI Act does provide some insight on this. Article 50 requires that AI content deemed to have been (possibly manipulated by) AI should be clearly labelled (possibly with some kind of tagging or watermark), so audiences are aware when they are being presented with synthetic media, especially when advertising or posting sponsored posts (European Parliament, 2023). But it is already apparent how there would be grey areas, what does “clearly” mean in this context, and how regulators are going to enforce these provisions, considering there is no accountability with international social platforms?


As lawmakers are working through these varied issues, the cash is rolling in. Brands are increasingly more interested in working with AI Influencers due to their ability to not age, never create a scandal, and can be created for any marketing intention. They can also monetize behind the scenes and cut deals for the content they have created in the form of sponsorship deals, licensing deals, virtual performances, and even digital merchandise. Lil Miquela has pulled in millions in brand partnerships, showing us that companies are willing to dump money, sometimes hundreds of thousands into influence that simply does not exist in the physical world (Economic Times, 2025).

The rise of AI influencers makes us rethink both online authenticity and what it means to be a “creator” and monetize it. If digital characters are getting contracts and creating followings, what does that mean for human influencers competing against each other for people’s time and money? Should we view AI influencers as tools, as businesses, or as new types of celebrities?

References:

Economic Times. (2025, September 7). Who are Mia Zelu and Lil Miquela? The rich, famous and fake influencers. The Economic Times. https://economictimes.indiatimes.com/news/international/us/who-are-mia-zelu-and-lil-miquela-the-rich-famous-and-fake-set-of-influencers/articleshow/123743591.cms

European Parliament. (2023, October). Generative AI: Watermarking and transparency requirements (EPRS Briefing). European Parliamentary Research Service. https://www.europarl.europa.eu/RegData/etudes/BRIE/2023/757583/EPRS_BRI(2023)757583_EN.pdf

Independent. (2025, July 9). Wimbledon AI influencer Mia Zelu confuses fans on Instagram. The Independent. https://www.independent.co.uk/life-style/wimbledon-ai-influencer-mia-zelu-instagram-b2787956.html

Lil Miquela [@lilmiquela]. (n.d.). Lil Miquela (official profile) [Instagram profile]. Instagram. Retrieved September 12, 2025, from https://www.instagram.com/lilmiquela/

Mia Zelu [@miazelu]. (n.d.). Mia Zelu (official profile) [Instagram profile]. Instagram. Retrieved September 12, 2025, from https://www.instagram.com/miazelu/?hl=en

Official Journal of the European Union. (2024). Regulation (EU) 2024/1689 of the European Parliament and of the Council on Artificial Intelligence (AI Act). https://artificialintelligenceact.eu/article/50/

Please rate this