Gemini products in Google Cloud, such as
Gemini Code Assist, incorporate responsible AI design principles
to help you navigate the capabilities, limitations, and risks of generative AI.
Understanding these safety controls and operational boundaries lets you evaluate
and deploy AI assistance across your workflows.

## Capabilities and risks of large language models

Large language models (LLMs) can perform many useful tasks such as the
following:

- Translate language.
- Summarize text.
- Generate code and creative writing.
- Power chatbots and virtual assistants.
- Complement search engines and recommendation systems.

At the same time, the evolving technical capabilities of LLMs create the
potential for misapplication, misuse, and unintended or unforeseen consequences.

LLMs can generate output that you don't expect, including text that's offensive,
insensitive, or factually incorrect. Because LLMs are incredibly versatile, it
can be difficult to predict exactly what kinds of unintended or unforeseen
outputs they might produce.

Given these risks and complexities, products such as
Gemini Code Assist are designed with
[Google's AI principles](https://ai.google/responsibility/principles/) in
mind. However, it's important for users to understand some of the limitations of
Gemini products to work safely and responsibly.

## Limitations of Gemini products in Google Cloud

Some of the limitations that you might encounter using Gemini products
in Google Cloud include (but aren't limited to) the following:

- **Edge cases.** Edge cases refer to unusual, rare, or exceptional situations
  that aren't well represented in the training data. These cases can lead to
  limitations in the output of Gemini models, such as model
  overconfidence, misinterpretation of context, or inappropriate outputs.

- **Model hallucinations, grounding, and factuality.** Gemini
  models might lack grounding and factuality in real-world knowledge, physical
  properties, or accurate understanding. This limitation can lead to model
  hallucinations, where Gemini might generate outputs that are
  plausible-sounding but factually incorrect, irrelevant, inappropriate, or
  nonsensical. Hallucinations can also include fabricating links to web pages
  that don't exist and have never existed. For more information, see
  [Write better prompts for Gemini in Google Cloud](https://docs.cloud.google.com/gemini/docs/discover/write-prompts.md.txt).

- **Data quality and tuning.** The quality, accuracy, and bias of the prompt
  data that's entered into Gemini products in Google Cloud can
  have a significant impact on its performance. If users enter inaccurate or
  incorrect prompts, Gemini might return suboptimal or false responses.

- **Bias amplification.** Language models can inadvertently amplify existing
  biases in their training data, leading to outputs that might further reinforce
  societal prejudices and unequal treatment of certain groups.

- **Language quality.** While Gemini products in Google Cloud
  yield impressive multilingual capabilities on the benchmarks that we
  evaluated against, the majority of our benchmarks (including all of the
  fairness evaluations) are in American English.

  Language models might provide inconsistent service quality to different users.
  For example, text generation might not be as effective for some dialects or
  language varieties because they are underrepresented in the training data.
  Performance might be worse for non-English languages or English language
  varieties with less representation.
- **Fairness benchmarks and subgroups.** Google Research's fairness analyses of
  Gemini models don't provide an exhaustive account of the various
  potential risks. For example, we focus on biases along gender, race,
  ethnicity, and religion axes, but perform the analysis only on the American
  English language data and model outputs.

- **Limited domain expertise.** Gemini models have been trained
  on Google Cloud technology, but it might lack the depth of knowledge
  that's required to provide accurate and detailed responses on highly
  specialized or technical topics, leading to superficial or incorrect
  information.

  When you use the **Gemini** pane in the Google Cloud console,
  Gemini is not context aware of your specific environment, so
  it cannot answer questions such as "When was the last time I created a VM?"

  In some cases, Gemini products in Google Cloud send a specific
  segment of your context to the model to receive a context-specific
  response---for example, when you click the **Troubleshooting suggestions**
  button in the Error Reporting service page.

## Gemini safety and toxicity filtering

Prompts and responses for Gemini products in Google Cloud,
such as Gemini Code Assist, are checked
against a comprehensive list of safety attributes as applicable for each use
case. These safety attributes aim to filter out content that violates our
[Acceptable Use Policy](https://cloud.google.com/terms/aup). If an output is
considered harmful, the response will be blocked.

## What's next

- Learn more about [how Gemini cites sources when helps you generate code](https://docs.cloud.google.com/gemini/docs/discover/works.md.txt#how-when-gemini-cites-sources).