<node id="692343">
  <nid>692343</nid>
  <type>news</type>
  <uid>
    <user id="27303"><![CDATA[27303]]></user>
  </uid>
  <created>1788533315</created>
  <changed>1788533640</changed>
  <title><![CDATA[Assessing Scientific Claims: Agent-Based AI System Answers Tough Questions]]></title>
  <body><![CDATA[<p>Advances in materials science and drug discovery increasingly depend on robotic laboratories that synthesize and test candidate compounds without human intervention. Artificial intelligence models can now propose custom materials and potential drug molecules far faster than robotic labs can physically test them. The resulting pools of candidate materials dramatically outstrip the availability of experimental testing.<br>&nbsp;</p><p>Deciding which candidates deserve scarce laboratory time is itself a hard problem. It requires weighing scientific plausibility against practical constraints — likely impact, expected effectiveness, and eventual manufacturability — considerations that are difficult to apply consistently across thousands of candidates.<br>&nbsp;</p><p>An artificial intelligence (AI) program developed at the Georgia Tech Research Institute (GTRI) could provide a new way to narrow that hypothesis pool down to a rate that robotic laboratories can actually test, and more broadly, to evaluate a wide range of scientific and technical claims that would otherwise overwhelm human assessment.<br>&nbsp;</p><p>Known as FARSCAPE, the program was developed to support a research project organized by the U.S. Defense Advanced Research Projects Agency (DARPA) to evaluate a broad range of feasibility questions. By breaking down challenging questions into smaller components that can be evaluated by a team of independent computer agents, FARSCAPE uses a “chain of thought” approach to provide answers in the form of probabilities. It then helps humans check the rationales for its assessment.<br>&nbsp;</p><p><a href="https://gtri.gatech.edu/newsroom/assessing-scientific-claims-agent-based-ai-system-answers-tough-questions">Read the full article</a> on the Georgia Tech Research Institute news page</p><p>&nbsp;</p>]]></body>
  <field_subtitle>
    <item>
      <value><![CDATA[]]></value>
    </item>
  </field_subtitle>
  <field_dateline>
    <item>
      <value>2026-09-04T00:00:00-04:00</value>
      <timezone><![CDATA[America/New_York]]></timezone>
    </item>
  </field_dateline>
  <field_summary_sentence>
    <item>
      <value><![CDATA[An artificial intelligence program developed at the Georgia Tech Research Institute could provide a new way to evaluate scientific claims.]]></value>
    </item>
  </field_summary_sentence>
  <field_summary>
    <item>
      <value><![CDATA[<p>FARSCAPE, an artificial intelligence (AI) program developed at the Georgia Tech Research Institute (GTRI) could provide a new way to evaluate a wide range of scientific and technical claims that would otherwise overwhelm human assessment.</p>]]></value>
    </item>
  </field_summary>
  <field_media>
          <item>
        <nid>
          <node id="681071">
            <nid>681071</nid>
            <type>image</type>
            <title><![CDATA[FARSCAPE provides a new way to evaluate a wide range of scientific and technical claims]]></title>
            <body><![CDATA[<p>FARSCAPE provides a new way to evaluate a wide range of scientific and technical claims. The chart shows the step-by-step process.</p>]]></body>
                          <field_image>
                <item>
                  <fid>265423</fid>
                  <filename><![CDATA[farscape_mockup.jpg]]></filename>
                  <filepath><![CDATA[/sites/default/files/2026/09/04/farscape_mockup.jpg]]></filepath>
                  <file_full_path><![CDATA[http://hg.gatech.edu//sites/default/files/2026/09/04/farscape_mockup.jpg]]></file_full_path>
                  <filemime>image/jpeg</filemime>
                  <image_740><![CDATA[]]></image_740>
                  <image_alt><![CDATA[User looking at FARSCAPE process diagram on a screen]]></image_alt>
                </item>
              </field_image>
            
                      </node>
        </nid>
      </item>
      </field_media>
  <field_contact_email>
    <item>
      <email><![CDATA[]]></email>
    </item>
  </field_contact_email>
  <field_location>
    <item>
      <value><![CDATA[]]></value>
    </item>
  </field_location>
  <field_contact>
    <item>
      <value><![CDATA[<p>gtri.media@gtri.gatech.edu</p>]]></value>
    </item>
  </field_contact>
  <field_sidebar>
    <item>
      <value><![CDATA[]]></value>
    </item>
  </field_sidebar>
  <field_boilerplate>
    <item>
      <nid><![CDATA[]]></nid>
    </item>
  </field_boilerplate>
  <!--  TO DO: correct to not conflate categories and news room topics  -->
  <!--  Disquisition: it's funny how I write these TODOs and then never
         revisit them. It's as though the act of writing the thing down frees me
         from the responsibility to actually solve the problem. But what can I
         say? There are more problems than there's time to solve.  -->
  <links_related> </links_related>
  <files> </files>
  <og_groups>
          <item>1188</item>
      </og_groups>
  <og_groups_both>
          <item>
        <![CDATA[Artificial Intelligence]]>
      </item>
      </og_groups_both>
  <field_categories>
          <item>
        <tid>194606</tid>
        <value><![CDATA[Artificial Intelligence]]></value>
      </item>
      </field_categories>
  <core_research_areas>
          <term tid="193655"><![CDATA[Artificial Intelligence at Georgia Tech]]></term>
      </core_research_areas>
  <field_news_room_topics>
      </field_news_room_topics>
  <links_related>
      </links_related>
  <files>
      </files>
  <og_groups>
          <item>1188</item>
      </og_groups>
  <og_groups_both>
          <item><![CDATA[Research Horizons]]></item>
      </og_groups_both>
  <field_keywords>
          <item>
        <tid>187915</tid>
        <value><![CDATA[go-researchnews]]></value>
      </item>
      </field_keywords>
  <field_userdata><![CDATA[]]></field_userdata>
</node>
