AI Features in Ashby

this guide provides an overview of the ai powered features in ashby and the best practices when using them enabling ai features in ashby you can navigate to admin > organization setup > opt in features to opt in and enable some or all of the following ai features in ashby docid\ gvifm7n4yba4cypjjxx4q docid\ bnuwme8ukhqyrd fzye8z ai generated fraud signal summaries as part of the docid\ xveiv6cjcfywty7538spl tool (enabled separately from the fraud detection tool) docid\ gfxouwywswcwhej65od8s the docid\ kfmxauorkzx6 ezbu32po within the report builder & ai assisted report builder substitution tokens that generate sourcing content for docid\ ggqq4vkh ynmutuxv43ha and docid\ p8ax 2lzye5qh7grothr6 for candidate emails filter assistant in docid zyxrhb2dhzqnqwl3ugx1 docid 0msgwzhfexbtwoagfyrar automatic reply status classification for docid\ b2gy2rhnnoylkbhwcwou4 docid\ kcboq9bn ndp 9i isj2b to provide a feedback summary on the candidates profile ai generated summaries for docid\ urozlu2ekwlbdhrvklkbo docid\ xveiv6cjcfywty7538spl docid\ r2shceuhaeo0adh bm5ar you can use the toggles beside each ai feature to activate or deactivate the features as needed click the links above to be taken to guides with more information on each feature if your organization would like to have future ai features enabled by default as they are released, you can switch on the auto enable new features option if your organization would prefer to review then enable new ai features manually, you can switch or leave this option off best practices below are some best practices to bear in mind when using the ai features in ashby ai content assistant once your content has been generated, review it fully if you’re happy to use that content moving forward, click looks good insert into text box alternatively, you can also ask another question if you’d like to give feedback on the generated content, click give feedback you’ll be able to provide a rating (positive or negative) and also add comments ai notetaker please note that the ai notetaker is currently in beta for more on consent management and data retention, check out docid\ bnuwme8ukhqyrd fzye8z of the ai notetaker guide looking to improve the suggestions generated for feedback form fields? check out our recommendations docid\ bnuwme8ukhqyrd fzye8z ai candidate assistant please note that the ai candidate assistant is still learning, so please review results when asking questions about candidates to confirm accuracy you can provide feedback on the assistant by clicking give feedback you’ll be able to add a rating along with any comments ai assisted report builder both ashby all in one and ashby analytics users can opt in to use this feature the ai report assistant is only available on the plus and enterprise plans foundations legacy plus plus enterprise ❌ ❌ ✅ ✅ please review all generated results carefully if the ai assisted report builder does not built the report you’re looking for from your prompt, you can click the give feedback button to provide a rating and add any comments ai generated email tokens when placing an ai generated email token in an email template or sequence, the token is context aware of its placement within the email body as such, it can be used within a sentence or to generate an entire sentence or paragraph within the email when drafting your email, the token will automatically start generating content in the body of the email once generation is complete, please review the content and either accept and add the generated content to the draft by clicking the checkmark icon regenerate the content using the regenerate icon remove the content and the token by clicking the trash can icon once you’ve taken action on the generated content, you will then be able to send the email to the candidate candidate search ai assistant both ashby all in one and ashby analytics users can opt in to use this feature when adding in a prompt to generate your candidate search filter, follow the description style listed in the examples underneath the describe your filter… field once your filter has been generated, please review accordingly to ensure the filters provided pull the information requested in your prompt you can use the thumbs up and thumbs down icons to provide feedback on the results provided once you’ve clicked either option, you can click tell us more… to provide feedback on the generated filter ai assisted application review click docid 0msgwzhfexbtwoagfyrar for more information on setting good job criteria please note, that any application form criteria you configure will only determine whether or not the question was answered and not the quality or accuracy of the answer blank or unrelated answers will be flagged as not meeting the criteria please make sure to review the application along with the criteria evaluation to determine the evaluation’s accuracy if any evaluations are inaccurate, you can flag them by clicking the three dots on the evaluation and then clicking flag evaluation automated reply status review any classifications that have been generated via ai to ensure accuracy all ai generated classifications will have an icon beside the reply status to indicate that has been classified by the feature you have the option to override the auto classification for up to one week after the classification is made click the response drop down to change the reply status as needed feedback summaries when summaries have been generated, be sure to check over all completed feedback forms to ensure the generated summary is an accurate representation of each reviewer’s viewpoint if you’d like to generate a new summary, click regenerate on the feedback summary within the candidate’s profile candidate review summaries when summaries have been generated, be sure to check over all completed reviews to ensure the generated summary is an accurate representation of each reviewer’s viewpoint candidate fraud detection when reviewing ai generated summaries, please also make sure to carefully review each signal too as you review each signal, you can hover over the signal title to get more information on what is being evaluated any flagged signals will be listed above the signals checked so you can easily see what needs checking once you’ve reviewed the data provided, you can choose from the following options to confirm the candidate’s fraud status mark fraud flag this as a fraudulent application this will then trigger the process to archive the candidate with the rejected by organization fraudulent candidate archive reason applied not fraud mark the flagging as a false positive and confirm the application is not fraudulent the candidate will have a fraud signals cleared label added to their profile unsure mark the application as uncertain flagged for monitoring so it can be re reviewed later if you need to update the fraud signals status on a candidate’s profile, you can click on the fraud signals label applied to the candidate’s profile to reopen the fraud signals history popover, then select the correct status ai report intepretation tool when generating your interpretation, you can open up each generated insight to get further information on your results if the insights don't match up or you're seeing an out of sync status for the intepretation, click regenerate to update the intepretation an interpretation status can show as out of sync when the filters or fields shown on a report have been amended