Google Gemini was launched in 2023, and is Google's spin on the rapidly developing AI virtual assistant technology. Because Google Gemini is built as a large language model (LLM), it can go far beyond what other AI platforms such as ChatGPT, or Microsoft Copilot are capable of (Micheal McDonough). It can be used not just as a personal helper, notetaker, and schedule maker, but it can also solve complex problems, thinking how humans would. Gemini is special because of its unique ability to not only process text, but also files, voices, and even videos, creating a one of a kind unique AI experience that no other platform can compare to.
Google Gemini stands out as a powerful AI assistant capable of understanding and generating text, images, audio, video, and even code (Pichai). A prominent feature is Audio Overview, which allows users to upload PDFs, DOCX, RTF, or TXT files and convert them into pod-cast style audio summaries narrated by ai generated voices (Crookes). A new feature with incredible potential is Gemini Live. This feature allows users to have real-time back and forth audible conversations with Gemini. A part of this feature also allows users to share their phone’s camera with Gemini to get real-time analysis of items around them. This feature has a wide variety of applications. For example, Gemini Live offers users the ability to receive real-time cooking guidance through spoken, camera-driven interactions. By simply holding down the power button, users can show the camera their ingredients or the dish they’re preparing, and Gemini will provide step-by-step instructions, troubleshooting tips (like fixing a soggy cake or bland curry), or recipe suggestions tailored to what’s on screen (El Khoury).
Google Gemini provides a wide range of services for academic, professional, and daily uses that can be useful at any level of education and profession. With the assistance of Google Gemini, not only can professors and teachers save time planning and creating lesson plans, but they can personalize their lessons to tailor to the needs of their students. Gemini has unique features that can create quiz questions from descriptions given, and find different ways to explain topics to cater to the different ways students learn. For students, it cannot only help facilitate learning by providing examples, visuals, and practice material, but it can also help with note-taking and assist in writing tasks.
The professional uses that Google Gemini offers can be greatly appreciated when using Google Suite, as Gemini can assist in writing emails and documents using Google Docs and other Google products. With the use of Google Gemini, companies can boost their productivity and communication with employees as administrators can use Gemini for implementing policies throughout their company, organizing schedules, meetings, and presentations. Any sort of professional can use Gemini for drafting and finalizing grant proposals, formal letters, reports, and research.
Google Gemini is an industry leading development that exceeded expectations when it was released, and while it is continuously updated, fixing bugs and expanding its capabilities, there are some limitations. One of these limitations is common within all AI platforms, nicknamed “hallucinations” (MIT Sloan) where the large language model (LLM) generates information that seems reliable and true, but is simply made up, producing misinformation in a convincing tone. This is an issue that exists across all AI platforms and is not unique solely to Google Gemini, however it is concerning and could cause problems especially in a professional or academic environment.
Another significant concern when it comes to not just AI platforms, but any online platform is privacy and data security. Since the dawn of social media recurring scandals involving illegal data brokering and collection have unfortunately been a reality. Google has aspirations to expand Gemini to be much more than a stand alone AI assistant, with goals to incorporate it into the google home system and as well as into your own personal Google account (Alastair Paterson). This raises a lot of red flags for some users who worry about their privacy and sensitive data being exposed. Google has taken note of these concerns about privacy and has assured users that they will abide by the “AI/ML privacy commitment which outlines our belief that customers should have the highest level of security and control over their data” (“How Gemini for Google Cloud”).
As Google Gemini and other Generative AI platforms expand their presence from fiction into our daily life, a number of ethical concerns have emerged. One prominent issue is the potential displacement of human labor. As Amazon CEO Andy Jassy acknowledged, AI “will lead to [a] smaller workforce” at Amazon (Jassy). This trend is not only seen at Amazon as the entire white-collar workforce has seen a collective 3.5% reduction (Cutter and Weber). These reductions raise concerns for the impacted peoples and their families.
On the flip side, the World Economic Forum estimates that while AI will disrupt 92 million jobs, it will also create 78 million new job opportunities by 2030 (World Economic Forum). This cycle can be said to parallel other historical cycles of technological disruptions that come from technological advancements, such as the destruction of the whaling industry with the advent of kerosene, petroleum, and electric lighting.
Within the academic world, AI raises ethical concerns around academic integrity and authorship. Students are increasingly using AI tools like Gemini to generate work to use as their own. A recent Guardian investigation found that nearly 7,000 students were proven to have cheated using AI during the 2023-24 school year representing 5.1 proven cases per 1,000 students, up by 3.5 cases from the previous year (Goodier). While cheating has always been an issue, AI allows students to do so at an unprecedented scale as AI generated work is almost indistinguishable from human generated work. The report states that the proven cases of academic dishonesty using AI are just the tip of the iceberg. These findings undermine core academic values, such as intellectual honesty and the integrity of earned degrees.
Honesty is not only being threatened by the humans using AI but the AI itself. There are valid concerns about generative AI’s reliability and fairness. The foundations of LLM’s (Large Language Models) are built using massive amounts of data to “train” different models. Bias and inaccurate data can be accidentally used within the training data leading to larger flaws with the finished product. Outside of bad training data, generative AI can also make up information and present it as fact, a phenomenon known as “hallucinations” (Burke). When users rely on these outputs without verifying them, it can lead to misinformation being repeated or believed, resulting in long-term consequences across education, research, and decision-making.
To help address these ethical concerns, Google has committed to improving Gemini with each new release. According to Google, Gemini models “undergo extensive ethics and safety tests” before deployment, with results integrated directly into their development and evaluation processes (Pichai). Each new version of Gemini, such as the shift from Gemini 1.0 to 1.5, incorporates new research and red-teaming techniques to uncover risks and reduce harm before those issues reach users (Pichai). This approach shows Google's recognition of Gemini’s ethical risks and their commitment to treating ethics as an ongoing responsibility.
In short, the development and release of Google Gemini has catapulted Generative AI backed by large language modeling to the front of tech research. With an already massive platform, and still growing, Google has put itself in a great position to push Gemini to be the most widely used AI digital assistant. Gemini’s image and video processing capabilities, as well as its human-like conversational attributes, and most importantly Google’s commitment to user privacy and data security have set Gemini apart from other AI platforms. Because widespread adoption of Generative AI into daily lives is still a newer development, research is constantly coming out about practicality, side effects, and long term outlook on AI. Ultimately, Google Gemini is a top contributor in the shift of how humans interact with technology on a daily basis.